Using mutt to access imap folders (was: mail questions: mutt and KDE)

[Giorgos Keramidas]

freebsd-kde stripped from the recipient list.

Gary it's probably better to post *two* messages if you have two
unrelated questions.  This way the traffic in freebsd-kde will not get
'polluted' with all the replies about mutt, which isn't really related
to KDE on FreeBSD.

On 2008-01-23 18:24, Gary Kline <[EMAIL PROTECTED]> wrote:
> Here's hopping that I'm still subscribed to the kde list, and that
> somebody can answer these general mail-type questions.

You don't have to be subscribed to freebsd-kde to post.

> Also, I would like some clues of how to if IF I can use mutt *with*
> IMAP.

You can set ${MAIL} in your environment to an imap folder.  Then firing
up mutt will select the imap folder by default, and you will be prompted
for a password.

Try for example the following:

$ MAIL='imap://[EMAIL PROTECTED]/INBOX'

You can store the IMAP username and password in ~/.muttrc too, but then
you should pay careful attention to the permissions of the file (even in
an internal network, it never hurts to be a bit cautious).  The
following in your ~/.muttrc should do the trick:

set imap_user='gkeramidas'
set imap_pass='achmed!the!terrorist'

Then fire up mutt, and enjoy your new IMAP session.

One word of caution though.  Setting $MAIL to an IMAP folder path in
your default environment may upset programs which don't really speak
IMAP the way mutt does, i.e. the command-line mail(1) utility.

The manual of mutt, which is accessible by hitting  while inside
mutt, is also a valuable resource for this sort of questions.
Specifically, the following section includes a very good example of how
you can handle multiple IMAP accounts in a single mutt session:

| 13. Managing multiple IMAP/POP accounts (OPTIONAL)
|
| If you happen to have accounts on multiple IMAP and/or POP
| servers, you may find managing all the authentication settings
| inconvenient and error-prone. The account-hook command may
| help. This hook works like folder-hook but is invoked whenever
| you access a remote mailbox (including inside the folder
| browser), not just when you open the mailbox.
|
| Some examples:
|
| account-hook . 'unset imap_user; unset imap_pass; unset tunnel'
| account-hook imap://host1/ 'set imap_user=me1 imap_pass=foo'
| account-hook imap://host2/ 'set tunnel="ssh host2 /usr/libexec/imapd"'

There are *many* other details in the manual about using mutt to access
IMAP folders.  You should definitely skim through the text at least
once in case you find some interesting option.

A text version of the full manual is available in /usr/local too:

/usr/local/share/doc/mutt/manual.txt

This is, actually, the file that you get to read when you hit 
inside mutt.

- Giorgos

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

Re: Sendmail local only.....

[Giorgos Keramidas]

On 2008-01-23 13:44, Agus <[EMAIL PROTECTED]> wrote:
>> Giorgos, question...is it needed to put the
>> sendmail_submit_enable="YES"
>> sendmail_msp_queue_enable="YES"
>> Although they are the default ones?
> 
> Maybe i wasnt clear...but with the options you gave me Giorgos i still
> can send email from localhost to external servers like hotmail for
> instance, through telnet.I want to disable this, so it can only
> send mails to local accounts

Ah, I misunderstood the original question.  My apologies.

On 2008-01-23 17:26, Matthew Seaman <[EMAIL PROTECTED]> wrote:
> Hmm... It's actually pretty hard to stop an MTA from being able to
> send e-mail, given that is what it is designed to do.

Hehe, indeed :)

> Another approach worth trying would be to use a wildcard mailertable
> entry that redirects any mail back to root the local machine:
> 
> 
>localhost  local:
>your.host.name local:
>.  local:root

That should work fine :)

> or replace the last line with:
> 
>.  error:Mail to external sites not allowed

That's a great idea to build upon.

>From the limited testing I did just now, bounces still go through, so
someone may abuse this to post email by setting the sender address to
the intended recipient, and posting to random outside mailboxes.

The following works too (after setting up a local dev-null alias):

localhost   local:
my.hostname local:
.   local:dev-null

This silently drops email messages, but at least it logs the delivery to
the dev-null alias, so there's a trace of the email message even though
it won't really get delivered to any user's mailbox and there's no error
responce sent to anyone.

Jan 24 03:57:09 kobe sendmail[9255]: m0O1ura8009255: [EMAIL PROTECTED], \
  size=83, class=0, nrcpts=1, msgid=<[EMAIL PROTECTED]>, \
  [EMAIL PROTECTED]
Jan 24 03:57:09 kobe sendmail[9255]: m0O1ura8009255: [EMAIL PROTECTED], \
  [EMAIL PROTECTED] (1000/1000), delay=00:00:16, xdelay=00:00:00, mailer=relay, 
\
  pri=30083, relay=[127.0.0.1] [127.0.0.1], dsn=2.0.0, \
  stat=Sent (m0O1v94R009303 Message accepted for delivery)
Jan 24 03:57:09 kobe smtpd[9303]: m0O1v94R009303: from=<[EMAIL PROTECTED]>, \
  size=374, class=0, nrcpts=1, msgid=<[EMAIL PROTECTED]>, \
  proto=ESMTP, daemon=Daemon0, relay=kobe.laptop [127.0.0.1]
Jan 24 03:57:09 kobe smtpd[9307]: m0O1v94R009303: to=/dev/null, \
  ctladdr=bit-bucket (26/0), delay=00:00:00, xdelay=00:00:00, mailer=*file*, \
  pri=30629, dsn=2.0.0, stat=Sent

Setting up a `mail-sink' alias, which traps all outgoing messages this
way, may be an easier way to recover the undelivered messages.  It
fringes on privacy violation, though, so I'm now going to pick one or
the other :)

- Giorgos

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

Re: sendmail config

[Giorgos Keramidas]

On 2008-01-23 19:54, Steven Friedrich <[EMAIL PROTECTED]> wrote:
> I configured KMail with SSL to send mail from a foreign network thru my 
> broadband ISP.  This works fine.
> 
> I want to be able to use send-pr, and thought I might me able to configure 
> sendmail as outgoing only. I have sendmail_enable="NO" in /etc/rc.conf.
> 
> local mail works fine.  I don't want incoming from the Internet.
> 
> I copied /etc/mail/freebsd.mc to laptop.mc and set SMART_HOST to  
> mail.insightbb.com and moved sendmail.cf to sendmailcf.sav.  I compiled 
> laptop.mc with m4 /usr/share/sendmail/cf/m4/cf.m4 laptop.mc
> 
> I then copied laptop.cf to sendmail.cf and rebooted.

There's a much easier way to generate `sendmail.cf' and `submit.cf' on
FreeBSD.  After reading the comments in `/etc/mail/Makefile', especially
the part shown below:

  # 
  # This Makefile uses `.mc' as the default MTA .mc file.  This
  # can be changed by defining SENDMAIL_MC in /etc/make.conf, e.g.:
  #
  #   SENDMAIL_MC=/etc/mail/myconfig.mc
  #
  # If '.mc' does not exist, it is created using 'freebsd.mc'
  # as a template.
  #
  # It also uses '.submit.mc' as the default mail submission .mc
  # file.  This can be changed by defining SENDMAIL_SUBMIT_MC in
  # /etc/make.conf, e.g.:
  #
  #   SENDMAIL_SUBMIT_MC=/etc/mail/mysubmit.mc
  #
  # If '.submit.mc' does not exist, it is created using
  # 'freebsd.submit.mc' as a template.
  # 

You can set up a `sendmail.cf' and `submit.cf' file by:

  0. Set up the `rc.conf' options for local delivery only, and to
 forward all other messages to the SMART_HOST relay.

sendmail_enable="NO"
sendmail_outbound_enable="NO"
sendmail_submit_enable="YES"
sendmail_msp_queue_enable="YES"

  1. Editing `/etc/make.conf' and setting SENDMAIL_MC and
 SENDMAIL_SUBMIT_MC to:

SENDMAIL_MC?= /etc/mail/laptop.mc
SENDMAIL_SUBMIT_MC?= /etc/mail/laptop.submit.mc

  2. Backup your current `laptop.mc' and `laptop.submit.mc' if any.

  3. Copy the original `freebsd.mc' to `laptop.mc' and the original
 `submit.mc' to `freebsd.submit.mc' in `/etc/mail':

# cd /etc/mail
# cp freebsd.mc laptop.mc
# cp freebsd.submit.mc laptop.submit.mc

  4. Edit the new `laptop.mc' file and set SMART_HOST.

  5. Use the standard `Makefile' to build and install the new *.cf files:

# cd /etc/mail
# make all && make install

Now you should be able to start Sendmail with:

# /etc/rc.d/sendmail stop
# /etc/rc.d/sendmail start

> Now, the response is "Service not available", but I think it's because
> I don't have sendmail configured to use SSL, like KMail.

The standard `freebsd.mc' distributed with FreeBSD sources doesn't force
the use of SSL or TLS.  You should be able to use it without SSL.

If you have made modifications to the original `freebsd.mc' file, you
can restore it by copying a fresh version of the same file from their
pristine copies included in the source tree of FreeBSD:

/usr/src/etc/sendmail/freebsd.mc
/usr/src/etc/sendmail/freebsd.submit.mc

HTH,
Giorgos

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

Re: Sendmail local only.....

[Giorgos Keramidas]

On 2008-01-22 16:28, Schiz0 <[EMAIL PROTECTED]> wrote:
>On Jan 22, 2008 4:25 PM, Agus <[EMAIL PROTECTED]> wrote:
>> Hi guys,
>> Just trying to make sendmail work locally only.And by that i
>> mean, i dont want sendmail to be able to relay or send mail to any
>> other machine except for localhost
>>
>> How can i do this? I mean is there a way on rc.conf or i need to
>> touch the cf??
>
> Search the file /etc/defaults/rc.conf for "sendmail"
> It has a bunch of options and explains what they do.
> NOTE: do not edit /etc/defaults/rc.conf - that's just the "Default
> settings" file.
> Make any changes in the /etc/rc.conf file instead.
>
> Basically, you want
>
> sendmail_enable="NO"
>
> And the rest of the sendmail options enabled.

Not all of them, mind you.  I use the following on my laptop:

sendmail_enable="NO"
sendmail_outbound_enable="NO"
sendmail_submit_enable="YES"
sendmail_msp_queue_enable="YES"

A fairly good explanation of why this setup works only as a 'local MTA',
and doesn't accept incoming SMTP connections from the world can be found
in the manpage of rc.sendmail:

% man rc.sendmail

To the original poster:

Feel free to use the above sendmail_xxx settings in your `/etc/rc.conf'
file, but please make sure that you also:

... understand why these settings make Sendmail only forward locally
generated email messages, by reading the rc.sendmail manpage.

... have configured properly your local `/etc/mail/sendmail.cf' and
`/etc/mail/submit.cf' files.

... read the Handbook section about Sendmail, and let us know if you
find it difficult to understand, or you would like to see more stuff
documented.

Cheers,
Giorgos

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

Re: Wrong times written by mkisofs?

[Giorgos Keramidas]

On 2008-01-20 11:16, Joerg Schilling <[EMAIL PROTECTED]> wrote:
>Bob Johnson <[EMAIL PROTECTED]> wrote:
>>> Let me asume you use a _recent_ mkisofs and set up a correct timezone...
>>>
>>[...]
>>>
>>> What do you get from "mkisofs -version"?
>>
>> mkisofs 2.01 (i386-unknown-freebsd6.2)
> 
> This is nearly 4 years old, why don't you use a recent version?
> recent is 2.01.01a37

FWIW,

2.01 is the version available in the Ports tree as `sysutils/cdrtools'.
The 2.01.01a37 version is available too, as `cdrtools-devel'.

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

Re: syslogd not reading messages from a remote machine

[Giorgos Keramidas]

On 2008-01-12 15:50, Jeffrey Goldberg <[EMAIL PROTECTED]> wrote:
> On Jan 11, 2008, at 9:51 AM, Andy Greenwood wrote:
>> I have recently set up a Fortigate-60 to run as a firewall/vpn on my
>> home network. I have a FreeBSD 7.0-prerelease machine sitting behind
>> it in the DMZ which is running ssh/web/etc. I'm trying to get the FG
>> to log to the BSD box's syslog. I have set up the necessary stuff on
>> the FG, and can send test logs from there to the bsd box. Running
>> tcpdump on the bsd [...]
>>
>> So I know that the packets are getting to the machine. I've set up
>> syslogd to accept packets from 10.10.10.1/32 in rc.conf, and
>> confirmed that the FG's IP should be accepted [...]
>>
>> I've restarted syslogd after every change I've made, but no dice. Can
>> anyone shed some light on why these messages aren't logging and what
>> I need to do to fix it?
>
> I'm sure that there is a simple answer for getting syslogd to work
> properly.  But after similar experiences to yours (on other systems), I now
> use syslog-ng (in ports) for any system that is going to be a remote syslog
> server.

To the original poster, since I missed the message which started the
thread...

Andy, the default mode of syslogd in FreeBSD is `secure mode', because
the -s option is enabled in `/etc/defaults/rc.conf':

% [EMAIL PROTECTED]:/root# grep '^syslogd_' /etc/defaults/rc.conf
% syslogd_enable="YES"# Run syslog daemon (or NO).
% syslogd_program="/usr/sbin/syslogd" # path to syslogd, if you want a 
different one.
% syslogd_flags="-s"  # Flags to syslogd (if enabled).
% [EMAIL PROTECTED]:/root#

The -s flag in the default options means that syslogd will *not* log
messages from remote hosts.  Quoting the syslogd manpage:

 -s  Operate in secure mode.  Do not log messages from
 remote machines.  If specified twice, no network
 socket will be opened at all, which also disables
 logging to remote machines.

I'm using the following in `/etc/rc.conf':

% [EMAIL PROTECTED]:/root# grep '^syslogd_' /etc/rc.conf
% syslogd_enable="YES"
% syslogd_flags=""
% [EMAIL PROTECTED]:/root#

and the syslogd on the system where this message is typed properly logs
messages from my DSL modem:

% [EMAIL PROTECTED]:/root# fgrep ' fire ' /var/log/messages | head -1
% Jan 12 22:04:26 fire kernel: Intrusion -> IN=ppp_8_35_1 OUT= MAC= \
% SRC=62.1.173.240 DST=62.1.60.253 LEN=48 TOS=0x00 PREC=0x00 \
% TTL=126 ID=39175 DF PROTO=TCP SPT=2213 DPT=445 WINDOW=16384 \
% RES=0x00 SYN URGP=0
% [EMAIL PROTECTED]:/root#

NOTE: If you are also using a firewall, you will have to make sure that
incoming UDP messages for the `syslog' service are allowed.  This is the
`pf.conf' snippet from my home network setup:

pass in proto udp from 192.169.1.0/24 to any port = syslog

HTH,
Giorgos

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

Re: Sendmail: "exposed" root, why?

[Giorgos Keramidas]

On 2008-01-08 13:26, "Philip M. Gollucci" <[EMAIL PROTECTED]> wrote:
> Jerahmy Pocott wrote:
>> Hello,
>>  From the sendmail documentation:
>> "There are always users that need to be "exposed" -- that is, their
>> internal site name should be displayed instead of the masquerade
>> name. Root is an example (which has been "exposed" by default prior
>> to 8.10)."

The `root' user is no longer exposed, so the really *imporant* question
is ``why are you still running Sendmail 8.10?''

> There is no directive, but you can edit the resultant .cf file
> and remove the line
> C{E}root
> or root from that line if more than one user.

The expose directives were part of the default OSTYPE and DOMAIN macros:

OSTYPE(freebsd6)
DOMAIN(generic)

> Be warned, you _will_ break /etc/crontab and periodic scripts mail
> delivery.

That's a possibility, yes :)

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

Re: My sendmail appears to be fixed, advice needed though

[Giorgos Keramidas]

On 2008-01-05 17:13, Andrew Falanga <[EMAIL PROTECTED]> wrote:
> Well, oddly enough I moved those additional lines to a position before the
> MAILER macros (I'll post the whitbap.mc file below as it exists now).
> However, I still got those error messages:
>
> Jan  5 18:29:10 whitbap sm-mta[6207]: NOQUEUE: SYSERR(root): opendaemonsocket:
> daemon MTA: cannot bind: Address already in use
> Jan  5 18:29:10 whitbap sm-mta[6207]: daemon MTA: problem creating SMTP socket
> Jan  5 18:29:15 whitbap sm-mta[6207]: NOQUEUE: SYSERR(root): opendaemonsocket:
> daemon MTA: cannot bind: Address already in use
>
> Anyway, I'm going to remove that line I mentioned before as this seems to make
> things work.  Please help me to resolve this and thanks very much for the
> help, this is great.

Ok, not can you show us the contents of `/etc/rc.conf' related to
Sendmail?  It's the next thing we have to check to ensure you are not
starting up multiple Sendmail listeners on the default smtp port.

% grep -i sendmail /etc/rc.conf

should do it :)

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

Re: How do I get sendmail working again

[Giorgos Keramidas]

On 2008-01-05 11:18, Andrew Falanga <[EMAIL PROTECTED]> wrote:
> Wanted to post the contents of my .mc file:
>
> whitbap# cat whitbap.mc
> [...]
> FEATURE(`no_default_msa')
> MAILER(local)
> MAILER(smtp)

This part seems ok.

> dnl set SASL options
> dnl TRUST_AUTH_MECH(`GSSAPI DIGEST-MD5 CRAM-MD5 LOGIN')dnl
> dnl define(`confAUTH_MECHANISMS', `GSSAPI DIGEST-MD5 CRAM-MD5 LOGIN')dnl
> define(`confAUTH_MECHANISMS',`PLAIN LOGIN')dnl
> TRUST_AUTH_MECH(`PLAIN LOGIN')dnl
> define(`CERT_DIR', `/etc/mail/certs')dnl
> define(`confCACERT_PATH', `CERT_DIR')dnl
> define(`confCACERT', `CERT_DIR/whitbap_cert.pem')dnl
> define(`confSERVER_CERT', `CERT_DIR/whitbap_cert.pem')dnl
> define(`confSERVER_KEY', `CERT_DIR/whitbap_key.pem')dnl
> define(`confCLIENT_CERT', `CERT_DIR/whitbap_cert.pem')dnl
> define(`confCLIENT_KEY', `CERT_DIR/whitbap_key.pem')dnl
> DAEMON_OPTIONS(`Port=smtp, Name=MTA')dnl
> DAEMON_OPTIONS(`Port=smtps, Name=TLSMTA')dnl

But I'm not sure it is a good idea to keep options like these *AFTER*
the MAILER() macros.  The `README' file of Sendmail's macros says:

% Beware: MAILER declarations should only be followed by LOCAL_*
% sections.  The general rules are that the order should be:
%
% VERSIONID
% OSTYPE
% DOMAIN
% FEATURE
% local macro definitions
% MAILER
% LOCAL_CONFIG
% LOCAL_RULE_*
% LOCAL_RULESETS
%
% There are a few exceptions to this rule.  Local macro definitions which
% influence a FEATURE() should be done before that feature.  For example,
% a define(`PROCMAIL_MAILER_PATH', ...) should be done before
% FEATURE(`local_procmail').

You can read the entire `README' file at:

/usr/share/sendmail/cf/README

But first give it a try, and move the last part of your *.mc file before
the MAILER() macro calls.

- Giorgos

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

Re: Paging Matthew Seaman

[Giorgos Keramidas]

On 2008-01-04 09:59, Paul Schmehl <[EMAIL PROTECTED]> wrote:
> I figure if anyone knows the answer to this off the top of their head,
> Matthew will.

I'm sure Matthew knows the answer, but you should use a more relevant
subject.  This way the archived posts will be more useful to future
posters.

> I've been reading the man pages for du and df, but I can't find the right
> combination.  I'd like to get the type of output that df -h gives you but
> only for one mount point or even one directory.  Is there a tool that can
> do that? (IOW, I'd like to run du -h but only get the totals for
> directories.)

Use the -s option of the `du' utility:

[EMAIL PROTECTED]:/home/keramida$ du -sk tmp/emacs-src
187436  tmp/emacs-src
[EMAIL PROTECTED]:/home/keramida$ du -sh tmp/emacs-src
183Mtmp/emacs-src
[EMAIL PROTECTED]:/home/keramida$

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

Re: Finally, Flash seems to be working for Me :)

[Giorgos Keramidas]

On 2008-01-04 00:42, Aryeh Friedman <[EMAIL PROTECTED]> wrote:
> On 1/4/08, Robert Huff <[EMAIL PROTECTED]> wrote:
> > Rudy writes:
> > >  I do do more than watch youtube videos, but people send me links
> > >  and I'm always bummed if I have to reboot into Ubuntu just to see
> > >  a stupid video  :)
> >
> > What you need is www/youtube-dl.  Deposits the content in a
> > form that can be played by mplayer (and presuably others).
> >
>
> I forget the exact name of it (away from my desktop machine) but there
> is an extension for firefox that wraps all video plugins to use an
> external player like mplayer

Lots of them, in fact.  The one I use is called "UnPlug".

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

Re: My sendmail appears to be fixed, advice needed though

[Giorgos Keramidas]

On 2008-01-05 13:08, Andrew Falanga <[EMAIL PROTECTED]> wrote:
> Hi everybody,
> Sorry for this flurry of e-mail from me over the last few days.  This
> has been highly frustrating.

You should post *more* details, not less.  One of the things which was
missing from the older posts (or at least, one thing which I didn't see)
was a *FULL* copy of your local *.mc configuration file.

> Basically, the instructions for setting up SSL and Authentication (from
> http://www.puresimplicity.net/~hemi/freebsd/sendmail.html) instruct to have
> the following lines in the .mc file:
>
> define(`confAUTH_MECHANISMS',`PLAIN LOGIN')dnl
> TRUST_AUTH_MECH(`PLAIN LOGIN')dnl
> define(`CERT_DIR', `/etc/mail/certs')dnl
> define(`confCACERT_PATH', `CERT_DIR')dnl
> define(`confCACERT', `CERT_DIR/mycert.pem')dnl
> define(`confSERVER_CERT', `CERT_DIR/mycert.pem')dnl
> define(`confSERVER_KEY', `CERT_DIR/mykey.pem')dnl
> define(`confCLIENT_CERT', `CERT_DIR/mycert.pem')dnl
> define(`confCLIENT_KEY', `CERT_DIR/mykey.pem')dnl
> DAEMON_OPTIONS(`Port=smtp, Name=MTA')dnl
> DAEMON_OPTIONS(`Port=smtps, Name=TLSMTA, M=s')dnl
>
> I deleated this line:
> DAEMON_OPTIONS(`Port=smtp, Name=MTA')dnl

That doesn't sound right.  I think it's because you have the options
listed above *after* the MAILER() calls.  This means that the options
are not `set' at the time the MAILER() calls generate your
configuration.  So you probably end up with several instances of the
`MTA' and `TLSMTA' daemon definitions in the final `sendmail.cf' file.

This could very well be the explanation of why your Sendmail *is*
listening on port :25 and it *also* tries to listen again, logging the
failures in syslog.

> Is this a viable fix or will I be missing something?  I was able to
> telnet to port 25 and send mail that way.  I cannot relay without
> logging in, so I think it's fixed.

See my previous post about moving MAILER() calls to the end of the file,
and give it another spin.  You are *very* close to figuring it all out
and having a fully working setup, AFAICT :)

Regards,
Giorgos

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

Re: corporate backers of freebsd

[Giorgos Keramidas]

On 2007-12-31 16:57, Colin Percival <[EMAIL PROTECTED]> wrote:
> Giorgos Keramidas wrote:
> > Yes, Gary, there are companies who also fund FreeBSD work in several
> > ways [...] Some examples which I recall off the top of my head are:
> 
> Don't forget pair Networks, which has generously supported phk, andre,
> and myself on our respective "sponsored FreeBSD coding" fundraising
> drives of 2004, 2005, and 2006, with slightly over $40,000 in total.

I knew I was forgetting something.  Thanks for the addition :-)

On 2007-12-31 18:00, Warren Block <[EMAIL PROTECTED]> wrote:
> On Tue, 1 Jan 2008, Giorgos Keramidas wrote:
>>  The support of Isilon Systems for VFS locking, which was then 'ported'
>>  back to FreeBSD.  Jeff Roberson worked with Isilon Systems to bring
>>  VFS locking to FreeBSD, and it is not part of the official kernel
>  ^^^
>>  source tree.
> 
> Probably s/not/now/ ...

That is true.  The VFS locking code *is* part of the source tree.
Unfortunately, spell-checking the original text couldn't really warn
me about this.  Thanks for this correction too :)

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

Re: corporate backers of freebsd

[Giorgos Keramidas]

On 2007-12-31 23:14, Pollywog <[EMAIL PROTECTED]> wrote:
>On Monday 31 December 2007 18:10:55 Gary Smithe wrote:
>> In short, here's my question:
>> Canonical, RedHat, IBM, Novell, and a slew of others are funding /
>> supporting Linux development and pushing some of that development
>> into the free community, so that all can benefit from full-time
>> developers and the money that supports them.
>
> This is true but on the flip side, Linux has become a target of the
> patent trolls (Microsoft and SCO to name two) and thus far they have
> left FreeBSD alone.  This was one thing that prompted me to try
> FreeBSD and to not depend exclusively on one OS solution.  I am just
> an individual user but this was still important to me.
>
> BTW you may have read that SCO has been delisted by NASDAQ   :)

This is probably a bit off-topic for the original thread, but SCO has
filed for bankruptcy.  This is probably related to being delisted by
NASDAQ.

To help a bit with the original question too:

Yes, Gary, there are companies who also fund FreeBSD work in several
ways, i.e.:

  * Paying developers on a per-project basis, to implement features
which they need.

  * Employing developers, and then contributing code back to the FreeBSD
Project.

  * Supporting FreeBSD drivers and subsystems, and working with the
Project to keep their hardware support up to date, to implement new
features, fix bugs, and so on.

Some examples which I recall off the top of my head are:

  The support of Isilon Systems for VFS locking, which was then 'ported'
  back to FreeBSD.  Jeff Roberson worked with Isilon Systems to bring
  VFS locking to FreeBSD, and it is not part of the official kernel
  source tree.

  NetApp and Isilon systems have made public statements, through the
  FreeBSD Foundation, about the reasons they like FreeBSD.  Advocacy of
  this sort, from successful companies is also a good contribution to
  the Project.  It may convince other companies to look at FreeBSD too.

  The hwpmc(4) performance counter work was started by Joseph Koshy, and
  then sponsored by Google and the FreeBSD foundation.

  NLNet supports the work of Marko Zec for the Network Stack
  Virtualization project.

  Cisco, iXsystems, Chelsio, Intel, Myricom, Neterion and others are
  actively contributing hardware to our "netperf" cluster.  Sentex is
  hosting the netperf cluster, and has been providing ongoing support to
  the FreeBSD Project for a very long time now.

  ISC is hosting FreeBSD Project machines too.

  Yahoo!, Apple, Juniper, Philips and Cisco are employers of some of the
  most active FreeBSD developers, and they have contributed in many many
  ways to the well-being and ongoing development of FreeBSD as we know
  it today.

  Cisco has provided, through Randall R. Stewart, a fully functional
  version of the SCTP protocol, and Randall has done an excellent job
  both of integrating SCTP into the tree, and supporting / maintaining
  it later on.

  Last, but definitely not least, Google, through its wonderful `Summer
  of Code' projects, has funded the development of a huge number of
  features which are either already part of the main FreeBSD source
  tree, or are in the process of being refined, debugged, tested and
  integrated to the main FreeBSD system.  The list of all the Google SoC
  projects is available through out web site, but here are some of the
  projects which I remember as I'm typing this:

* BSD bintools project (some of the tools which are part of the GNU
  binutils have been cleanly implemented using only BSD-licensed
  code).

* Improvements to the Ports infrastructure.  This was completed and
  committed to CVS by Gabor Kovesdan, who was fudned by Google for
  _two_ years in a row.  Kudos to both Gabor and Google for all the
  Ports work they have done :)

* SNMP monitoring and a BSD-licensed snmpd daemon has been
  implemented by Shteryana Shopova and committed to the tree.

This is, by far, not an exchaustive list, but just a *few* of the
companies which have supported the FreeBSD Project so far.  I have
undoubtedly forgot many more, since I am both a relatively "new" FreeBSD
team member, and I am not involved in *all* the sub-projects which are
part of the greater FreeBSD Project "umbrella".

There is a lot more information on our web site about companies who
contribute to the development of FreeBSD.  The quarterly status reports
at  and the `Newsflash' at
 are good places to "hunt"
for this sort of information.

HTH,
Giorgos

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

Re: Building FreeBSD from source on OS X?

[Giorgos Keramidas]

On 2007-12-31 16:33, "Johan A. van Zanten" <[EMAIL PROTECTED]> wrote:
> Greetings.
> Recently i acquired an intel-based Mac mini and would like to build
> FreeBSD on it, but the mini needs to be (always) running OS X, because
> of other duties (DVR) it has.
> 
> Is it possible to build FreeBSD from source on OS X running on intel
> hardware?
> 
> I've tried the straight forward "make" in the source area, but it
> appears that OS X's make is actually gnu make, and it doesn't like the
> FreeBSD makefiles. I have the NetBSD pkgsrc system setup, and have a
> "bmake" which gets a little further when trying to build
> /usr/src/usr/bin/make, but still has some problems.
> 
> Before i go any further with this, i wanted to check and see if this
> is even possible.

I think one of the best options right now is to use Parallels, or some
other virtualization system.

BTW, I'm trying to 'port' FreeBSD make to make it buildable with
autoconf, and automake, so that it is easier to run freebsd-make on
non-FreeBSD systems, but there's still a lot of work before this is
usable for building a full FreeBSD source tree.

- Giorgos

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

Re: dlopen(), atexit() on FreeBSD

[Giorgos Keramidas]

On 2007-12-30 18:49, Markus Hoenicka <[EMAIL PROTECTED]> wrote:
> Hi,
> I bumped into a platform-specific problem when using the Firebird
> database client library in a dlopen()ed module on FreeBSD. libdbi
> (http://libdbi.sourceforge.net) is a database abstraction layer which
> dlopen()s available database drivers at runtime to provide
> connectivity to various database engines. This design works without
> problems on a variety of platforms and with a variety of database
> client libraries, but causes a segfault with Firebird on FreeBSD:
> 
> #0  0x28514fe4 in ?? ()
> #1  0x281507c3 in __cxa_finalize () from /lib/libc.so.6
> #2  0x281503fe in exit () from /lib/libc.so.6
> #3  0x0804a40f in main (argc=1, argv=0xbfbfe754) at test_dbi.c:419
> 
> The application crashes when exit() is called. Googling told me that
> __cxa_finalize () is invoked by atexit(). Our drivers and apps do not
> use this function, but the firebird client libraries do:
> 
> [EMAIL PROTECTED]:~/prog/libdbi-drivers/tests# grep atexit 
> /usr/local/lib/libfb*
> Binary file /usr/local/lib/libfbclient.so matches
> Binary file /usr/local/lib/libfbembed.so matches
> 
> Googling also told me that the conflict between atexit() and dlopen()
> on FreeBSD is a known problem, see e.g.:
> 
> http://www.imagemagick.org/pipermail/magick-developers/2006-March/002523.html
> 
> Is there anything I can do about this from my end?

The __cxa_finalize() function is not called by atexit(), but by exit()
itself.  I don't know of any way to `unregister' exit handlers, so one
way of `fixing' this is to avoid calling dlclose() before exit() in the
example code shown at the URL above:

% #include 
% #include 
%
% int main( int argc, char ** argv)
% {
% void * handle;
% const char * mod = "/usr/local/lib/libMagick.so";
% void (*InitializeMagick)(const char*);
%
% handle = dlopen( mod, RTLD_LAZY);
% if ( !handle) {
% fprintf( stderr, "cannot load %s\n", mod);
% exit(1);
% }
%
% (void*) InitializeMagick = dlsym( handle, "InitializeMagick");
% if ( !InitializeMagick) {
% fprintf( stderr, "cannot resolve InitializeMagick\n");
% exit(1);
% }
%
% InitializeMagick( "moo");
% dlclose( handle);
% }

Since the program is going to exit and have all its dlopened shared
objects be unmapped, it's probably ok to skip the dlclose() step in
this example.

In the case of the firebird libraries, since they are libraries things
are a bit trickier, because the consumers of these libraries don't
really know that a dlopened object has called atexit() :(

I think this is probably something that the freebsd-hackers list will
be interested in.  Can you post a description of the problem there too?

- Giorgos

___
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 make sendmail listen on an address other than the loopback

[Giorgos Keramidas]

On 2007-12-28 10:33, Andrew Falanga <[EMAIL PROTECTED]> wrote:
> Hi,
> I'm trying to get an e-mail system working for my church
> (whitneybaptist.org).  I've added a file called local-host-names in
> /etc/mail as described in the Handbook, then did "/etc/rc.d/sendmail
> restart" and then did "sockstat | grep sendmail" and got the following
> results:
> 
> root sendmail   32889 3  tcp4   127.0.0.1:25  *:*
> root sendmail   32889 4  dgram  -> /var/run/logpriv
> smmspsendmail   696   3  dgram  -> /var/run/log
> 
> Now, with the exception of the additional file, nothing has been done
> to this stock sendmail configuration (system is 6.2-RELEASE-p7).  How
> would I make sendmail listen on the ip of 192.168.2.23?  I do have
> some experience with sendmail, however, it was several years ago and
> I've forgotten quite a bit.  Why isn't it listening on that address
> now?

What you see is `normal' for a host which supports local email delivery
and forwards everything else to another `smart host'.  If you want to
start a listener which also accepts email from the network (instead of a
listener only for 127.0.0.1 like the one you have now), you will have to
tweak the `sendmail_xxx_enable' options in your `/etc/rc.conf' file.

Right now, you probably have something like:

  sendmail_enable="NO"
  sendmail_submit_enable="YES"
  sendmail_msp_queue_enable="YES"

To run the sendmail daemon for inbound email connections on all network
interfaces, you need sendmail_enable="YES", and then you can drop the
line for the `submit' daemon:

  sendmail_enable="YES"
  sendmail_msp_queue_enable="YES"

Note: For more details about these `rc.conf' variables, it may be useful
to read the rc.sendmail(5) manpage.

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

Re: removing ipfw rules

[Giorgos Keramidas]

On 2007-12-27 15:47, Noah <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I have two ipfw rules that I want to remove.  They are viewable with the
> "ipfw show" command
>
>
> --- snip ---
>
> 06600  0 0 allow ip from any to any proto tcp src-ip
> 66.66.66.66 dst-port 22
> 06700  0 0 allow ip from any to any proto tcp src-ip
> 66.66.66.66 dst-port 22
>
> --- snip 
>
> I am typing the command "/sbin/ipfw -q delete pass proto tcp src-ip
> 66.66.66.66 dst-port 22"
>
> but both lines remain.  What am I doing wrong?

There are differences between the visible rule:

allow ip from any to any proto tcp src-ip 66.66.66.66 dst-port 22

and the one you are trying to delete:

pass proto tcp src-ip 66.66.66.66 dst-port 22

Having said that, can you try something simpler, i.e.

ipfw -q delete 6600
ipfw -q delete 6700

This should work too, if I remember well enough the ipfw syntax.

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

Re: mutt??

[Giorgos Keramidas]

On 2007-12-27 11:02, Gary Kline <[EMAIL PROTECTED]> wrote:
> I've been trtryinng to rebuild everything on tao to get my i810
> graphics working. Somehow, mutt bbroke. It seems to break with
> something undefined in perl5.8.
>
> Anybody know what this is:
> Undefined symbol "__sbmaskrune"  ?

That's odd.  The mutt-devel port (which I am using to type and post this
message) does not seem to depend on Perl:

% $ pkg_info -r mutt\*
% Information for mutt-devel-...
%
% Depends on:
% Dependency: ispell-3.2.06_18
% Dependency: mime-support-3.39.1
% Dependency: libiconv-1.11_1
% Dependency: gettext-0.16.1_3
%
% $

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

Re: Buildworld RELENG_7 problem

[Giorgos Keramidas]

On 2007-12-26 06:34, Bernt Hansson <[EMAIL PROTECTED]> wrote:
> Hello!
> 
> I'v got a bit of a problem upgrading from FreeBSD 7.0-BETA2 #0
> to any later version.

> This is the error I get:
> 
> > gtyp-gen.h
> echo "NULL};"   >> gtyp-gen.h
> echo "static const char * const lang_dir_names[] = {"   >> gtyp-gen.h
> echo "\"c\", "  >> gtyp-gen.h
> echo "\"cp\", " >> gtyp-gen.h
> echo "\"objc\", "   >> 
> gtyp-gen.h
> echo "NULL};"   >> gtyp-gen.h
> cc -O2 -fno-strict-aliasing -pipe -I. -DIN_GCC -DHAVE_CONFIG_H 
> -DPREFIX=\"/usr\" -I/usr/obj/usr/sr
> c/gnu/usr.bin/cc/cc_tools/../cc_tools 
> -I/usr/src/gnu/usr.bin/cc/cc_tools/../cc_tools -I/usr/src/gn
> u/usr.bin/cc/cc_tools/../../../../contrib/gcc 
> -I/usr/src/gnu/usr.bin/cc/cc_tools/../../../../contr
> ib/gcc/config 
> -I/usr/src/gnu/usr.bin/cc/cc_tools/../../../../contrib/gcclibs/include 
> -I/usr/src/gn
> u/usr.bin/cc/cc_tools/../../../../contrib/gcclibs/libcpp/include 
> -I/usr/src/gnu/usr.bin/cc/cc_tool
> s/../../../../contrib/gcclibs/libdecnumber -g -DGENERATOR_FILE 
> -DHAVE_CONFIG_H  -I/usr/obj/usr/src
> /tmp/legacy/usr/include -c 
> /usr/src/gnu/usr.bin/cc/cc_tools/../../../../contrib/gcc/gengtype.c
> In file included from ./tm.h:4,
>  from 
> /usr/src/gnu/usr.bin/cc/cc_tools/../../../../contrib/gcc/gengtype.c:24:
> ./options.h:901: error: redeclaration of enumerator 'OPT_w'
> ./options.h:899: error: previous definition of 'OPT_w' was here
> *** Error code 1

Hi Bernt,

Are you using a non-POSIX locale (i.e. LANG, LC_ALL and friends)?

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

Re: how long does send-pr take to post

[Giorgos Keramidas]

On 2007-12-20 17:19, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
> In the past I have not been able to do send-pr but now that I fixed my
> local mail issues all other email apps work... how long should I wait
> for the pr to show up before I decide some kind of error happened.

A few hours for the email notice saying 'we received your report, and it
has been filed as number XXX' is probably reasonable.

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

Re: Bash script to find out the summary of user memory usage [not working]

[Giorgos Keramidas]

On 2007-12-17 06:00, Patrick Dung <[EMAIL PROTECTED]> wrote:
> I have correction with the script but still doesn't work:
>
> #!/usr/local/bin/bash
> for user in `ps -A -o user | sort | uniq | tail +2`
>  do
> echo "user: $user"
>
>ps aux -U $user | tail +2 | while read line
>do
>
> mem=`echo $line | awk {'print $4'}`
> echo "mem: $mem"
> TMPSUMMEM=`awk -v x=$mem -v y=$TMPSUMMEM 'BEGIN{printf
> "%.2f\n",x+y}'`
> echo "summem: $TMPSUMMEM"
>done
> echo "finalsummem: $SUMMEM"
> export SUMMEM=$TMPSUMMEM
>  done
>
> echo "finalsummem: $SUMMEM"

There are *many* race conditions in that script.  For example, there's
no guarantee that once you get a snapshot of the "ps -A -o user" output,
then the same users will be listed in the loop you are running for each
username.

The script is also a bit 'sub-optimal' because it calls ps(1) and parses
its output many times (at least as many times as there are users).  A
much better way to `design' something like this would be to keep a hash
of the usernames, and keep incrementing the hash entry for each user as
you hit ps(1) output lines.

I'm not going to even bother writing a script to use a hash in bash(1),
because there are much better languages to work with hashes,
dictionaries or even simple arrays.

Here's for example a Python script which does what I described:

 1  #!/usr/bin/env python
 2
 3  import os
 4  import re
 5  import sys
 6
 7  try:
 8  input = os.popen('ps xauwww', 'r')
 9  except:
10  print "Cannot open pipe for ps(1) output"
11  sys.exit(1)
12
13  # Start with an empty dictionary.
14  stats = {}
15
16  # Regexp to strip the ps(1) output header.
17  header = re.compile('USER')
18
19  for line in input.readlines():
20  if header.match(line):
21  continue
22  fields = line.split()
23  if not fields or len(fields) < 4:
24  continue
25
26  (username, mem) = (fields[0], float(fields[3]))
27  value = None
28  try:
29  value = stats[username]
30  except KeyError:
31  pass
32
33  if not value:
34  stats[username] = 0.0
35  stats[username] += mem
36
37  # Print all the stats we have collected so far.
38  keys = stats.keys()
39  if len(keys) > 0:
40  total = 0.0
41  print "%-15s %5s" % ('USERNAME', 'MEM%')
42  for k in stats.keys():
43  print "%-15s %5.2f" % (k, stats[k])
44  total += stats[k]
45  # Finally print a grand total of all users.
46  print "%-15s %5.2f" % ('TOTAL', total)

It's not the shortest Python script one could write to do what you
describe, but I've gone for readability rather than speed or
conciseness.

Running this script should produce:

$ ./foo.py
USERNAME MEM%
_pflogd  0.10
daemon   0.00
bind 1.10
_dhcp0.10
keramida38.60
smmsp0.10
root10.10
build0.00
TOTAL   50.10
$

PS: Yes, you could probably do the same in bash, with sed, awk and a bit
of superglue, but I prefer Perl and/or Python for anything which
involves something a bit more involved than simple string substitution
these days...

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

Re: Cannot get Script to Run Via Crontab

[Giorgos Keramidas]

On 2007-12-17 07:19, Robert Huff <[EMAIL PROTECTED]> wrote:
>Giorgos Keramidas writes:
>>> The lines from my script that are causing the problem are:
>>>[...]
>>> The cron message to mail/root ends with:
>>>
>>>exec: ps2pdf12: not found
>>>
>>> I am assuming that cron cannot find a path or a config file for
>>> ghostscript, but I don't have any idea how to fix this problem.
>>  
>>  Yes.  That's what is happenning.  The default PATH of cron jobs doesn't
>>  include `/usr/local/bin', but you have lots of options:
>>  
>>1) Add it to the crontab file
>>  
>>2) Modify the default path in your Perl script:
> 
> Allow me to recommend the second, as it will not disturb other cron
> programs that may be expecting the default path.

That is a very good point :)

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

Re: Cannot get Script to Run Via Crontab

[Giorgos Keramidas]

On 2007-12-16 19:10, David Goodnature <[EMAIL PROTECTED]> wrote:
> I have a perl script that I can execute from the command line as root.
> It runs fine.  When I try to automate it using the root crontab, the
> script fails.
>
> The lines from my script that are causing the problem are:
>
>my $scomd = "/usr/local/bin/ps2pdf -dPDFSETTINGS=/prepress 
> -dProcessColorModel=/DeviceGray -dAutoRotatePages=/PageByPage 
> -dDownsampleMonoImages=true -dMonoImageDownsampleType=/Average 
> -dMonoImageDownsampleThreshold=1.5 -dMonoImageResolution=600 
> ".$inpath.$cur_ps_files[0]." ".$outpath.$pdffilename;
>
>### create the new .pdf file from the .ps file
>system($scomd) == 0 or return "system $scomd failed: $?";
>
> The cron message to mail/root ends with:
>
>exec: ps2pdf12: not found
>
> I am assuming that cron cannot find a path or a config file for
> ghostscript, but I don't have any idea how to fix this problem.

Yes.  That's what is happenning.  The default PATH of cron jobs doesn't
include `/usr/local/bin', but you have lots of options:

  1) Add it to the crontab file

PATH="/bin:/sbin:/usr/bin:/usr/sbin:/usr/local/bin:/usr/local/sbin"

crontab entries here

  2) Modify the default path in your Perl script:

$ENV{PATH} = 
'/bin:/sbin:/usr/bin:/usr/sbin:/usr/local/bin:/usr/local/sbin';

my $scomd = join(' ', ('/usr/local/bin/ps2pdf',
 '-dPDFSETTINGS=/prepress',
 '-dProcessColorModel=/DeviceGray',
 '-dAutoRotatePages=/PageByPage',
 '-dDownsampleMonoImages=true',
 '-dMonoImageDownsampleType=/Average',
 '-dMonoImageDownsampleThreshold=1.5',
 '-dMonoImageResolution=600'
 $inpath.$cur_ps_files[0],
 $outpath.$pdffilename));

system($scomd) == 0 or return "system $scomd failed: $?";

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

Re: Apparently, csh programming is considered harmful.

[Giorgos Keramidas]

On 2007-12-16 19:36, Chuck Robey <[EMAIL PROTECTED]> wrote:
> Michael P. Soulier wrote:
>> On 14/12/07 Giorgos Keramidas said:
>>> Tcsh is a fine shell.  I'm using it all the time (that's how I found out
>>> that a buglet reported by Kris Kennaway a few months ago was indeed a
>>> bug which I could reproduce too).
>> I always found csh/tcsh aliases annoying, since there are no shell 
>> functions.
>> I also found the shell redirection awkward. 
> 
> There;s one item that is much more easily done in csh/tcsh than in the sh 
> based ones  that's redirecting the stderr along with the stdout.  with 
> tcsh, when I do a make, I commonly do a:
> 
> make |& tee makeout
> 
> which causes both the stdout and stderr files to be redirected to the
> "makeout" make listing file.  I;'ve never figured out any reasonably
> simple way to do that in any sh-like shell.  Is there any simble way
> that you know of?

Yep, there is a simple way in sh too:

make 2>&1 | tee makeout

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

Re: pdksh vs. mksh info [was: Re: Apparently, csh programming is considered harmful.]

[Giorgos Keramidas]

On 2007-12-15 13:54, Frank Shute <[EMAIL PROTECTED]> wrote:
>> % [EMAIL PROTECTED]:/usr/local/bin$ ls -ld mksh bash ksh
>> % -rwxr-xr-x  1 root  wheel  -  684699 Dec  9 19:51 bash
>> % -r-xr-xr-x  1 root  wheel  - 2390645 Aug 31 17:07 ksh
>> % -r-xr-xr-x  1 root  wheel  -  236202 Dec  9 18:34 mksh
> 
> Wow. My pdksh is much smaller:
> 
> $ ls -ld /usr/local/bin/ksh
> -r-xr-xr-x  1 root  wheel  681584 Jan 23  2007 /usr/local/bin/ksh
> 
> $ pkg_info -W /usr/local/bin/ksh
> /usr/local/bin/ksh was installed by package pdksh-5.2.14p2_2
> 
> It's also statically compiled. I wonder what is bloating yours so
> much. Was it built with debugging code or something?

Yes.  All my ports are build with DEBUG_FLAGS='-g' this time, so it may
be the cause of the pdksh bloat.

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

Re: GOOD boook!

[Giorgos Keramidas]

On 2007-12-14 14:44, Gary Kline <[EMAIL PROTECTED]> wrote:
> Many thank to whomever recommended SED AND AWK a few weeks ago.
> Was it you, Giorgos?

Not really, no.  At least, my `sent-mail' folder didn't come up with any
matches.  But it's a jewel of a book, you're right about that...

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

pdksh vs. mksh info [was: Re: Apparently, csh programming is considered harmful.]

[Giorgos Keramidas]

On 2007-12-14 21:10, Frank Shute <[EMAIL PROTECTED]> wrote:
> I used bash for an interactive shell for about 5 years until I
> discovered the goodness of pdksh. About half the size, statically
> linked, not full of bugs and better editing features. Plus it's not
> GPL.

Hi Frank,

Now that you mention pdksh, have you tried mksh (in Ports too)?

I've installed it and successfully run moderately large ksh scripts
(like the webrev(1) utility of OpenSolaris), and it is about an order of
magnitude smaller than pdksh here:

% [EMAIL PROTECTED]:/usr/local/bin$ ls -ld mksh bash ksh
% -rwxr-xr-x  1 root  wheel  -  684699 Dec  9 19:51 bash
% -r-xr-xr-x  1 root  wheel  - 2390645 Aug 31 17:07 ksh
% -r-xr-xr-x  1 root  wheel  -  236202 Dec  9 18:34 mksh
% [EMAIL PROTECTED]:/usr/local/bin$ ldd mksh bash ksh
% mksh:
% libc.so.7 => /lib/libc.so.7 (0x280ae000)
% bash:
% libncurses.so.7 => /lib/libncurses.so.7 (0x28101000)
% libintl.so.8 => /usr/local/lib/libintl.so.8 (0x28144000)
% libiconv.so.3 => /usr/local/lib/libiconv.so.3 (0x28156000)
% libc.so.7 => /lib/libc.so.7 (0x2824b000)
% ldd: ksh: not a dynamic executable
% [EMAIL PROTECTED]:/usr/local/bin$

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

Re: Absolute FreeBSD

[Giorgos Keramidas]

On 2007-12-14 15:22, Barnaby Scott <[EMAIL PROTECTED]> wrote:
> I have just finished it and I would say it does exactly what what Ted
> and cpghost suggests it should - there are plenty of sections where
> the author introduces what can be done with a particular tool or part
> of the OS, and suggests to the reader to investigate further options
> in the approriate manuals. It also quite openly acknowledges that
> there is plenty that is not covered at all.
> 
> As someone with very limited experience (I'm not sure if I still
> classify as a *complete* newbie) I found the book an excellent and
> even entertaining read, which serves it purpose extremely well: to
> give an overview and introduction, but with enough detail in relevant
> places to be able to get real, useful stuff done.

Michael has a writing style which I like a lot, but I haven't had a
chance to read the second version yet.

I've read the first version cover to back, however, and it was written
in a very entertaining, elegant style.  Judging from my experience with
the first edition, I expect nothing less than what you just described :)

- Giorgos

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

Re: Apparently, csh programming is considered harmful.

[Giorgos Keramidas]

On 2007-12-13 21:59, Chuck Robey <[EMAIL PROTECTED]> wrote:
>Giorgos Keramidas wrote:
>> On 2007-12-13 18:05, Chad Perrin <[EMAIL PROTECTED]> wrote:
>>> I ran across this today:
>>>
>>>   http://www.faqs.org/faqs/unix-faq/shell/csh-whynot/
>>>
>>> Title:
>>>   Csh Programming Considered Harmful
>
> That was written sometime last millenium, I mean, it's REALLY old.  The
> question is sort of flamebait (you ought to go ask it on, say, the Linux
> IRC channel, for well-reasoned, adult discussion (NOT!)  In general, it's
> right, you really wouldn't want to use tcsh as a scripting language.  Read
> it, you'll come out ahead, but understand, that tcsh is a heck of a good
> general purpose command shell for users.

Tcsh is a fine shell.  I'm using it all the time (that's how I found out
that a buglet reported by Kris Kennaway a few months ago was indeed a
bug which I could reproduce too).

I just don't like it for scripting :-)

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

Re: Apparently, csh programming is considered harmful.

[Giorgos Keramidas]

On 2007-12-13 18:05, Chad Perrin <[EMAIL PROTECTED]> wrote:
> I ran across this today:
> 
>   http://www.faqs.org/faqs/unix-faq/shell/csh-whynot/
> 
> Title:
>   Csh Programming Considered Harmful
> 
> I wonder what responses I might get here, and how much of this applies
> to tcsh as well (I'm still not exactly a tcsh expert).

Most of the points made in the FAQ about scripting large `applications'
with csh ring a bell for me.  Now, having said that, /bin/sh is nice for
small to medium-sized scripts, but there is a certain point where even
sh(1) becomes annoying.

Do you have any _particular_ parts of the csh-whynot article that you
would like to discuss, or this is a free for all flame? :)

___
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 cut or awk commands into sed command ?

[Giorgos Keramidas]

On 2007-12-13 09:35, Halid Faith <[EMAIL PROTECTED]> wrote:
> Let me try to explain
> I have a file called A which contains variable values as below;
> file1, abc12
> foot1, cba11
> boby, def123
> ...
> 
> Also I have another file called B which contains partly valuable values as
> following;
> ### file of A begin
> Server valuable1
> Client valuable2
>  the file end
> 
> I have to assign the first column valuables in A to valuable1 in B and
> assign second column valuable1 in A to valuable2 in B.
>
> Finally I should see as following in a file called C
> Server file1
> Client abc12
> 
> Server foot1
> Client cba11
> 
> Server boby
> Client def123
>
> How can I do that ? Could you give me a script ?

Look carefully at the second input file:

Server valuable1
Client valuable2

Now look *very* carefully at the output you want:

Server file1
Client abc12

Server foot1
Client cba11

Server boby
Client def123

Can you imagine a *loop* which reads the input file and creates the
second?  In `pseudo-code' this would be something like:

for each pair of (server, client) from `file1':
for each line in `file2':
substitute `server' and `client' in `line'
print resulting line
print an empty line

That shouldn't be *too* hard to write in a shell script.  Now that you
have the outline, see if you have better luck.

- Giorgos

___
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 cut or awk commands into sed command ?

[Giorgos Keramidas]

On 2007-12-12 23:19, Halid Faith <[EMAIL PROTECTED]> wrote:
> I have a file named file1 which contains some values.
> I want to replace some strings into it, so I use sed command but I get an 
> error.
>
> sed "s#oldstring#`cut -d, -f3 file2`#"  file1
> sed: 1: "s/yenidomain2/f0b2875d- ...": unterminated substitute in regular 
> expression

That's not enough information to help you in a meaningful manner.

* What does `file2' have to do with the replacement strings, and why do
  you use it?

* What are the contents of both files?

* What do you want to replace, and what should it be replaced with?

> also I get an error with awk command into sed;
> sed "s#oldstring#`awk -F, '{print$3}' file2`#"file1
> sed: 1: "s#yenidomain2#f0b2875d- ...": unterminated substitute in regular 
> expression

That's not very different from the cut-based command.

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

Re: smart_host on sendmail min config

[Giorgos Keramidas]

On 2007-12-11 13:53, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
>Giorgos Keramidas wrote:
>> You still haven't shown us:
>>
>> * How your local rc system starts Sendmail
> 
> Until I can get it right from the command line I am not going to
> include it in my /etc/rc thus currently there is none
>
>> * The active Sendmail processes you have now
> 
> Except for the one I start (with the below command line) before
> sending there is none running.

Then you need to replicate what `/etc/rc.d/sendmail' does.  You have the
code to that script, you also have `/etc/defaults/rc.conf' as a guide of
the defaults supported by FreeBSD.  Happy hacking :)

I'm not interested in rewriting `/etc/rc.d/sendmail' from scratch, so
other than a few pointers like the following, I'm out of the thread:

> here is the complete transcript of my last attempt to actually send
> something: [...]

> monster-freebsd# sendmail -bd -q1m

Here you started a sendmail process which listens for connections to
*all* interfaces.  That's probably ok if you plan to accept incoming
SMTP connections on _any_ interface, i.e. because you are running a
Sendmail server accessible from outside.

On my laptop this would be replaced by:

/usr/sbin/sendmail -L smtpd -bd -q30m \
-ODaemonPortOptions=Addr=localhost

Note the use of the -O flag to make sure Sendmail is only listening on
the localhost:25 port for mail submission requests.

Then you have to start a *second* Sendmail instance, to flush the local
`clientmqueue' and forward the messages you see as `Deferred' when you
run the `mailq -Ac' command:

> monster-freebsd# mailq -Ac
> /var/spool/clientmqueue (8 requests)
> - -Q-ID- --Size-- -Q-Time-
> - Sender/Recipient---
> lBAEQDUM0772289 Mon Dec 10 09:26 aryeh
>  (Deferred: Connection refused by [127.0.0.1])
>  [EMAIL PROTECTED]

Now you have to run a second Sendmail instance, i.e. with:

/usr/sbin/sendmail -L mailq -Ac -q10m

NOTE: The `smtpd' and `mailq' arguments to the -L options are not the
standard names FreeBSD assigns to these processes.  They are just the
ones I use on my laptop.

Since you are using a custom, local sendmail startup script, that's
about the best I can write to get you kickstarted.  For more help with
the way Sendmail operates, you can always refer to the ``Sendmail(TM)
Installation and Operation Guide''.  A copy of the guide is conveniently
installed by the FreeBSD installation process at:

/usr/share/doc/smm/08.sendmailop/*

Good luck with the startup scripts.  I don't really understand why you
insist on rewriting something which works(TM), is featureful and easy to
extend, like the rc.d scripts, but that's besides the point :-)

- Giorgos

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

Re: smart_host on sendmail min config

[Giorgos Keramidas]

On 2007-12-11 13:25, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
>Giorgos Keramidas wrote:
>> Neat!  Your local Sendmail submission service works then.  You
>> should be able to see the message in the `clientmqueue' with:
>>
>> mailq -Ac
>>
>> Now, can you show us how your local rc system starts Sendmail, and
>> the active Sendmail processes you are using?  It may be working
>> fine, but only queueing messages in `/var/spool/clientmqueue'.  A
>> queue runner should also be started to periodically scan and handle
>> the clientmqueue messages.

I'm sorry, but this is the *third* time I am asking the same thing, and
it's getting a bit boring.  Do you _really_ want help, or just to have a
bit of email fun for a while?

You still haven't shown us:

* How your local rc system starts Sendmail

* The active Sendmail processes you have now

> monster-freebsd# !sen
> sendmail -bd -q1m -v
> # very long pause
> 050 WARNING: local host name (monster-freebsd) is not qualified; see
> cf/README: WHO AM I?

Your DNS is broken, and Sendmail cannot resolve `monster-freebsd'.
That's not a Sendmail issue, so I'm ignoring that for now.

> monster-freebsd# mailq -Ac
> # very long pause
> /var/spool/clientmqueue (8 requests)
> - -Q-ID- --Size-- -Q-Time-
> - Sender/Recipient---
> lBAEQDUM0772289 Mon Dec 10 09:26 aryeh
>  (Deferred: Connection refused by [127.0.0.1])
>  [EMAIL PROTECTED]

No submission service running.  You have to start at least some process
which listens on 127.0.0.1:25 for incoming email requests and a queue
runner for clientmqueue.  I haven't seen any "ps xauwww" output for the
Sendmail processes you are currently running, so no idea why you don't
have a listener on 127.0.0.1:25.

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

Re: smart_host on sendmail min config

[Giorgos Keramidas]

On 2007-12-11 13:02, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
>> Ideally, something like `mail -v' should show at least an attempt to
>> post the message to the local queue: [...]
>
> monster-freebsd# sendmail -bd -q1m -v
> 050 WARNING: local host name (monster-freebsd) is not qualified; see
> cf/README: WHO AM I?
> monster-freebsd# mail -v [EMAIL PROTECTED]
> Subject: test fopo
> dsadad
> .
> EOT
> WARNING: local host name (monster-freebsd) is not qualified; see cf/README: 
> WHO AM I?
> [EMAIL PROTECTED] Connecting to [127.0.0.1] via relay...
> 220 monster-freebsd ESMTP Sendmail 8.14.2/8.14.2; Tue, 11 Dec 2007
> 13:00:27 -0500 (EST)
> >>> EHLO monster-freebsd
> 250-monster-freebsd Hello localhost [127.0.0.1], pleased to meet you
[...]
> 354 Enter mail, end with "." on a line by itself
> >>> .
> 250 2.0.0 lBBI0RgU000340 Message accepted for delivery
> [EMAIL PROTECTED] Sent (lBBI0RgU000340 Message accepted for delivery)
> Closing connection to [127.0.0.1]
> >>> QUIT
> 221 2.0.0 monster-freebsd closing connection

Neat!  Your local Sendmail submission service works then.  You should be
able to see the message in the `clientmqueue' with:

mailq -Ac

Now, can you show us how your local rc system starts Sendmail, and the
active Sendmail processes you are using?  It may be working fine, but
only queueing messages in `/var/spool/clientmqueue'.  A queue runner
should also be started to periodically scan and handle the clientmqueue
messages.

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

Re: smart_host on sendmail min config

[Giorgos Keramidas]

On 2007-12-10 22:00, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
> Dec  9 09:15:38  newsyslog[587]: logfile first created
> Dec  9 09:15:38  sm-mta[720]: starting daemon (8.14.2): [EMAIL 
> PROTECTED]:30:00
> Dec  9 09:15:38  sm-msp-queue[724]: starting daemon (8.14.2): [EMAIL 
> PROTECTED]:30:00
> Dec  9 12:22:28  sm-mta[729]: starting daemon (8.14.2): [EMAIL 
> PROTECTED]:30:00
> Dec  9 12:22:28  sm-msp-queue[733]: starting daemon (8.14.2): [EMAIL 
> PROTECTED]:30:00
> 
> completely unchanged before and after I attempted to sendmail to my
> self (after a reboot but not a restart of sendmail [I have a
> handcrafted /etc/rc so I know it doesn't run sendmail on boot])

I don't see message sending attempts above.  Since you have modified
`/etc/rc' can you show me the processes launched for Sendmail, the way
they started, and then how you are trying to send an email message?

Ideally, something like `mail -v' should show at least an attempt to
post the message to the local queue:

% $ mail -v keramida
% Subject: foo
% testing
% .
% EOT
% keramida... Connecting to [127.0.0.1] via relay...
% 220 kobe.laptop ESMTP Sendmail 8.14.2/8.14.2; Tue, 11 Dec 2007 18:26:12 +0200 
(EET)
% >>> EHLO kobe.laptop
% 250-kobe.laptop Hello kobe.laptop [127.0.0.1], pleased to meet you
% 250-ENHANCEDSTATUSCODES
% 250-PIPELINING
% 250-8BITMIME
% 250-SIZE
% 250-DSN
% 250-ETRN
% 250-AUTH GSSAPI DIGEST-MD5 CRAM-MD5
% 250-DELIVERBY
% 250 HELP
% >>> MAIL From:<[EMAIL PROTECTED]> SIZE=35 [EMAIL PROTECTED]
% 250 2.1.0 <[EMAIL PROTECTED]>... Sender ok
% >>> RCPT To:<[EMAIL PROTECTED]>
% >>> DATA
% 250 2.1.5 <[EMAIL PROTECTED]>... Recipient ok
% 354 Enter mail, end with "." on a line by itself
% >>> .
% 250 2.0.0 lBBGQCj8018567 Message accepted for delivery
% keramida... Sent (lBBGQCj8018567 Message accepted for delivery)
% Closing connection to [127.0.0.1]
% >>> QUIT
% 221 2.0.0 kobe.laptop closing connection
% $

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

Re: smart_host on sendmail min config

[Giorgos Keramidas]

On 2007-12-10 15:08, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
>Chuck Swiger wrote:
>>On Dec 10, 2007, at 11:40 AM, Aryeh Friedman wrote:
>>> What else do I need to add to this to make it work (i.e. send all
>>> mail via mx1.optonline.net):
>>>
>>> OSTYPE(`freebsd6')dnl define(`SMART_HOST',
>>> `mx2.optonline.net')dnl
>>
>> Look at /etc/mail/freebsd.mc and edit the SMART_HOST line as above.
>> (Note that you can copy freebsd.mc to hostname.mc and use the latter
>> filename in order to avoid modifying the generic template file.)
>
> Just tried that the ISP just eats a test message (to the addr I am
> sending this from) but a direct telnet to port 25 on the smart host
> works fine

If a direct telnet to port 25 works, then SMART_HOST should work too.

Can you grab log messages from `/var/log/maillog', and post them?

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

Re: FreeBSD make on other platforms [was: Re: Building FreeBSD onLinux]

[Giorgos Keramidas]

On 2007-12-04 08:29, "Saravanan Shanmugham (sarvi)" <[EMAIL PROTECTED]> wrote:
> Hi Giorgos,
>  The build failed for me 'with a syntax error before "__dead2" '.
>  Are there any special include files I am supposed to have?
> I had trouble with the autoconf version I had installed and had to
> upgrade to the latest.
> Then I was able to move past, autoconf, automake, configure.

The project is in its 'prenatal' stage yet, but I'm trying to move a bit
faster towards a buildable source tree on Linux.  As Philip notes:

On 2007-12-04 11:33, "Philip M. Gollucci" <[EMAIL PROTECTED]> wrote:
> __dead2 is a custom gcc modification.  It just means the function
> doesn't return IIRC.

There's still a fair amount of work to make the source tree buildable on
non-FreeBSD platforms.  One of the next steps should be to make these
FreeBSD-specific bits (mostly stuff from src/sys/sys/cdefs.h) available
outside of the full FreeBSD src/ tree too.

Happily, the FreeBSD source is BSD-licensed, so I can import parts of it
to the bmake source tree easily :)

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

Re: Building FreeBSD on Linux

[Giorgos Keramidas]

On 2007-12-02 20:13, "Saravanan Shanmugham (sarvi)" <[EMAIL PROTECTED]> wrote:
> I installed/bootstrapped NetBSD pkgsrc on the linux box. This comes
> with bmake(which I think standas for bsdmake).

pkgsrc is a very different ``beast'' altogether.

It knows enough to bootstrap itself on any host with a vaguely
POSIX-compatible shell, and some sort of host-make(1) tool.

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

FreeBSD make on other platforms [was: Re: Building FreeBSD on Linux]

[Giorgos Keramidas]

On 2007-12-03 08:08, Micha?l Gr?newald <[EMAIL PROTECTED]> wrote:
>"Saravanan Shanmugham (sarvi)" <[EMAIL PROTECTED]> writes:
>> I have tried GNU Make 3.80 as well as pmake. And I can't seem
>> to find bmake for Linux.
>
> I have found in my vaults a script that downloads pmake source from
> your favorite FreeBSD's mirror and builds a binary that worked on (at
> least one workstation running) Linux.

Since there seems to be so much interest in running FreeBSD
make(1) on other platforms, I've committed what I have done so
far to .  It's not a full port
of FreeBSD make(1) yet, but anyone interested to help is invited
to hop in and give me a hand :)

The main goal of this `project' is to have a fully buildable
version of FreeBSD make(1) on Linux and Solaris, so that one can
build FreeBSD-like makefiles by running:

/usr/local/bin/bmake [ -f BSDmakefile ]

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

Re: Building FreeBSD on Linux

[Giorgos Keramidas]

On 2007-11-30 16:06, "Saravanan Shanmugham (sarvi)" <[EMAIL PROTECTED]> wrote:
>  
> Hi,
> I am trying to build all of FreeBSD from a Linux Machine and seem to
> be running into problems.  We have farm of build machines that we use
> to build many other things and my team would look like to use it going
> forward for our FreeBSD development.
>  
> Has anyone tried this before? 
>  
> I have tried GNU Make 3.80 as well as pmake. And I can't seem to find
> bmake for Linux.

I don't know of any port of BSD make(1) to Linux, or if that would be
sufficient to cross-build FreeBSD.  I'm trying to build a snapshot of
FreeBSD make(1) which builds with autoconf, for other stuff, but it
may take a while before I have a fully autoconf-ified version and that
may still not be adequate.

You can always try booting a FreeBSD snapshot inside vmware or qemu,
and build FreeBSD from FreeBSD, but that may be a bit slow...

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

Re: Strange kernel log message

[Giorgos Keramidas]

On 2007-11-26 09:58, Ceri Davies <[EMAIL PROTECTED]> wrote:
> So I have this in my security run output:
> kernel log messages:
> +++ /tmp/security.hLYJI0kFSun Nov 25 03:01:02 2007
> +222>2>>>NNNMNMMIII M  III SIISAS SAAA  3 303,020,0 ,,  EE IEIIESSSAIAA S 
> A f ff
> +
> +
> +f
>
> WTF now?
>
> I'm not sure if that's a real kernel message that got garbled or whether
> I should be worried about naughtiness.

It looks like multiple messages overlapping each other.  Removing 3
characters every 4 bytes in the output produces things which seem
vaguely recognizable:

<22NNI II A ,,,EISA  fff
<<2>NMI  SS 300  ISAAfff

There's a sysctl option which you can tweak to make this less likely to
happen, but I am not sure about its name.  Our console gurus can help
you track it down and tune its value :)

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

Re: Getting around ISP SMTP firewall settings (Re: Submitting a new port if send-pr is broken)

[Giorgos Keramidas]

On 2007-11-26 04:00, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
>BTW I a redirected this to -questions
>> You should be able to set up a local mailer/MTA (sendmail, postfix,
>> etc.) and tell it to use your ISP's mail server on TCP port 25, and
>> it all should just "magically work" unless they require SMTP AUTH
>> (not many do from what I've seen; they base authentication on the
>> source IP of customers).
>>
>> sendmail refers to this feature as SMART_HOST, while postfix refers
>> to it as a transport destination (see transport(5)).
>
> I have not set the MTA up yet for it but I did test it with
> thunderbird... an other question how can I set it up that I can
> receive mail (dynamic IP and 25 inbound is blocked)?

Thunderbird doesn't necessarily go through an SMTP connection to the
local host, so it may work with or without a local MTA installation &
setup (depending on which host you forward outgoing email).

If you set up Thunderbird to use `localhost' for outgoing email, then
you have to also configure a local MTA (Sendmail, Postfix, or qmail are
popular choices).

I don't think there's an easy way to set up the local Sendmail
installation to *receive* email from the world without some sort of
`static address' though.  To do that, you would have to work with your
ISP, so that:

* Your address does not change semi-randomly or ramdonly.

* Your fully qualified domain resolves correctly and its MX records
  point to your static IP address.

* Your incoming port 25 traffic is not filtered.

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

Re: top posting (off-topic)

[Giorgos Keramidas]

On 2007-11-25 19:01, Chad Perrin <[EMAIL PROTECTED]> wrote:
> On Sun, Nov 25, 2007 at 06:56:15PM -0700, Chad Perrin wrote:
> >
> > I think it's kind of a chicken-and-egg problem: we don't really know for
> > sure whether TOFU[1] posting spurred much of the rise of illiteracy or
> > the increase of relative illiteracy on the Internet led to an increase in
> > TOFU posting.  Which came first?
>
> I forgot to include the footnote about TOFU in the preceding message.  It
> would have looked something like this:
>
> [1]: TOFU = Text Over, Fullquote Under; the most common format of top
> posted replies

The footnote was easy to understand after a quick Wikipedia search:
http://en.wikipedia.org/wiki/Top-posting#Top-posting

Quoting the text (so list members don't have to actually repeat the
search):

Some maintain that top-posting is _never_ appropriate, and refer to
it jokingly as the "TOFU" method (from the German "text oben,
fullquote unten", sometimes translated "text over, fullquote
under") [...]

Nice one.  I had not heard of "TOFU posting" before :)

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

Re: who wrote this

[Giorgos Keramidas]

On 2007-11-25 19:43, "eBoundHost: Artur" <[EMAIL PROTECTED]> wrote:
>> If you are not bothered by the fact that the fortune cookie database
>> contains Hitler quotes, but you merely want to ammend the text of the
>> web site, then you are more than welcome to post patches to the
>> freebsd-www list.  That's where most of the work on the website is
>> discussed, and reviewed
>
> I'm not going to reply to your personal attack but will only say that you
> are definitely much smarter, more web savy and better looking than me.

It's not an ``attack'', but merely a request to avoid what may be
considered as a ``fault'' of the FreeBSD Project.

> As far as the rest of your comment, yes, I am only interested in
> removing the reference to hitler from the front page of that topic
> that's all.  And unfortunately I do not know how to post patches to
> the freebsd-www list so if you would be so kind as to send me an email
> with a pointer, I would sincerely appreciate it.

You don't have to spend a lot of time working with HTML, SGML or the CVS
tree.  A mere proposal for a text that improves what you consider buggy
in the web page, should be fine.  We can do the rest.

- Giorgos

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

Re: who wrote this

[Giorgos Keramidas]

On 2007-11-25 17:59, "eBoundHost: Artur" <[EMAIL PROTECTED]> wrote:
> yea that's a great answer.  thanks for your insight.  this is not some
> technical question that can be researched, this in fact tarnishes the
> image of the freebsd community, so it's not such an easy "go rtfm"
> type of deal.

The commit logs of the entire FreeBSD source tree are openly visible to
everyone.  A couple of minutes with Google Groups would be enough to
locate this thread:

  
http://groups.google.com/group/fa.freebsd.cvs-all/browse_thread/thread/ee66a0ebc1457b1f/7ea396d642a58806

This is the sort of "research" that I wanted to point you at.

> problem is that i just came accross it myself and obviously nothing
> has been done about it in the past.

The fact that you didn't notice the old commit logs, does not mean that
``nothing has been done about this issue in the past''.  Please stop
spreading FUD about FreeBSD, because implying that we don't care as a
team about these things suggests to readers of freebsd-questions things
which are untrue.

There are intriguing but nevertheless interesting observations in the
thread mentioned above.  Please take the time to let some of them sink
in before you spread more FUD.

> so i would like to ask of people, is there no better way to get the
> point accross?  do you have to have this wording?  is it set in stone
> and can't be changed?  I insist strongly that we should rework this
> example, and if anyone insists strongly on not doing it, I would like
> to understand what motive can be possibly behind this other than
> something very deeply evil.

If you are not bothered by the fact that the fortune cookie database
contains Hitler quotes, but you merely want to ammend the text of the
web site, then you are more than welcome to post patches to the
freebsd-www list.  That's where most of the work on the website is
discussed, and reviewed.

- Giorgos

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

Re: who wrote this

[Giorgos Keramidas]

On 2007-11-25 15:23, "eBoundHost: Artur" <[EMAIL PROTECTED]> wrote:
> All I have to say is WTF is wrong with whomever wrote this page.
> http://www.freebsd.org/internal/fortunes.html
> ==
>>> Examples of entries that should not usually be declared 'offensive':
>>> * Hitler quotes.
> ==
> Ok I understand that some moron wrote it, but why has nobody removed
> this garbage?

There have been long and vicious discussions about this particular
issue, on several mailing lists.  Please refer to the archives, because
repeating the whole shebang is not really a productive use of everyone's
time.

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

Re: Help with a new port?

[Giorgos Keramidas]

On 2007-11-24 15:24, Zachary Kline <[EMAIL PROTECTED]> wrote:
> Hi,
> My name is Zachary Kline,

Hi Zachary,

> Anyway, to get to the point: I'm not quite sure where to ask this.  I
> have a new port which I feel should be included in the FreeBSD
> accessibility category, [...]
> This port is Emacspeak, from http://emacspeak.sf.net.  It's a screen
> reader--though that term isn't really encouraged by the developer--for
> the Emacs work environment.

I can help with the integration of the new port.  I will have a look at
the site of the program, but it would be nice if you sent me any porting
details/work you have already.

Happy FreeBSD'ing :-)

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

Re: top posting (off-topic)

[Giorgos Keramidas]

On 2007-11-23 21:58, David Kelly <[EMAIL PROTECTED]> wrote:
>On Nov 22, 2007, at 9:10 PM, Paul Schmehl wrote:
>> Understood from that perspective, perhaps you can see why people
>> might dislike top posting.
>
> Many here (and elsewhere) will not reply to a top-poster.

I am one of these people.

If I see a top-posted message -- totally incomprehensible, full of
errors, misformattings, and other annoying bits, including mutilated
quotes with completely messed up quoting, and semi-randomly wrapped text
-- then it instantly rings a very important bell:

"The author of this message does not care enough to put some effort into
writing a properly formatted, readable reply.  If he doesn't care enough
to make his message readable, do you really want to spend the effort to
_read_ it?"

The answer is, surprisingly often, "No, I don't think I want to do that".

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

Re: can you help about this script

[Giorgos Keramidas]

On 2007-11-22 10:10, ann kok <[EMAIL PROTECTED]> wrote:
> Hi Giorgos
>
> Thank you
>
> But my output is from your suggstion
> printf "Created: %s\n", system("date +%Y%m%d");
>
> 20071122
> Created: 0
> 20071122
> Updated: 0
>
> how can I have output as
>
> Created: 20071122
> Updated: 20071122

You'll have to use the gsub() to strip newlines from the output of
"date"...

> In additon,
>
> ls it possible to have loop output also?
>
> I need to have
>
> print "File No:", CMA001
>
> the second record is CMA002 and then CMA003 for the
> 3rd record

Sure.  One way to do this is to print a formatted version of the special
"NR" variable of awk (NR == number of records read so far):

$ ( echo foo ; echo bar ) | awk '{ printf "%03d %s\n", NR, $0; }'
001 foo
002 bar
$

If you are going to do any amount of *serious* awk programming, I
recommend the following book:

Dale Dougherty, Arnold Robbins.  "Sed & Awk".  O'Reilly &
Associates.  2nd edition (March 1997)

http://www.oreilly.com/catalog/sed2/

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

Re: can you help about this script

[Giorgos Keramidas]

On 2007-11-21 12:26, ann kok <[EMAIL PROTECTED]> wrote:
> Hi all
> how command "date, hostname" run in awk program?
>
> awk -F program.awk file.txt

You don't use backticks...  These are a feature of the shell, and
running a script through progname.awk is no longer a shell session.

Try system("date") in your awk(1) script:

> program.awk
>
>  BEGIN { RS = "\n" ; FS = "|" }
>
>  {
>print "Name:", $9
>print "Created: `date`"   
>print "from: `hostname`"
>print ""
>  }


BEGIN {
RS ="\n";
FS = "|";
}

{
printf "Name:%s\n", $9;
printf "Created: %s\n", system("date");
printf "From:%s\n", system("hostname");
}

Running system("hostname") once for each file may be horribly
inefficient, though.  If I were you, I'd write this as a *shell* script,
which runs "hostname" once, stashes the result away in a variable, and
reuses it all the time.

Running "date" may be a bit less efficient than something like
gettimeofday().  Perl has a gettimeofday() function in the Time::HiRes
module, so it may be worth investigating if that may speed things up a
bit more.

A completely untested first try to do something like this is ...

#!/usr/bin/perl -w

use strict;

use POSIX qw(strftime);
use Time::HiRes qw(gettimeofday);

my $hostname = `hostname`;
my $line;
while (defined($line = )) {
chomp $line;
my @fields = split /|/, $line;
if ($#fields >= 0) {
my ($seconds, $microseconds) = gettimeofday();
printf "Name:%s\n", $fields[8];
printf "Created: %s\n",
strftime("%Y-%m-%d %H:%M:%S", gmtime($seconds));
printf "From:%s\n", $hostname;
}
}

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

Re: hyperthreading CPU and broken scheduling?

[Giorgos Keramidas]

On 2007-11-18 11:43, Wojciech Puchar <[EMAIL PROTECTED]> wrote:
>>> http://lists.freebsd.org/mailman/listinfo/freebsd-questions
>>> To unsubscribe, send any mail to 
>>> "[EMAIL PROTECTED]"
>> To enable hyperthreading, try setting the following in /etc/sysctl.conf:
>>
>> machdep.hyperthreading_allowed=1
>>
>> and reboot (or execute sysctl machdep.hyperthreading_allowed=1 by hand).
>
> did /etc/rc.d/sysctl start  and works.

Please, use loader.conf to change `machdep.hyperthreading_allowed'.

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

Re: congrlations to the freebsd developers

[Giorgos Keramidas]

On 2007-11-17 22:44, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
>Giorgos Keramidas wrote:
>> On 2007-11-17 18:53, "Aryeh M. Friedman" <[EMAIL PROTECTED]>
>> wrote:
>>> M. Warner Losh wrote:
>>>> You can run FreeBSD/i386 on amd64 boxes.
>>> All the different methods people have suggested to me for doing
>>> this has blown up also since I use a nvidia card it is almost
>>> pointless unless I can run the nvidia kernel module.
>>
>> That's odd.
>>
>> The laptop I'm typing this is amd64-capable, but installing from
>> the FreeBSD/i386 release CD-ROMs worked like a charm.  Which
>> methods have you tried?
>
> With 4GB and a E6850 (P35/IHC9(R)) almost all the pci devices I had
> where reconized *EXCEPT* any nvdidia card under i386 this was
> instantly "fixed" as soon I switched to amd64.

Sorry, but I can barely parse the sentence above.  Do you mean that you
_did_ installl i386, but it had issues?  Or that you had an _amd64_
installation, but was forced to move back to i386?

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

Re: Making mergemaster skip certain files

[Giorgos Keramidas]

On 2007-11-17 20:06, Howard Goldstein <[EMAIL PROTECTED]> wrote:
> Giorgos Keramidas wrote:
> > On 2007-11-16 22:34, "J. Porter Clark" <[EMAIL PROTECTED]> wrote:
> > mergemaster for more details, but here's a short description of what I
> > use on my laptop for some time now:
>
> > 1. A `.mergemasterrc' file in the HOME directory of the `root' user,
> >which contains:
> >[snip]
> >
> > 2. The `/root/mm-pre-compare.sh' script contains the following:
> >[snip]
>
> Please consider submitting this as a PR in hopes it gets included in
> the actual package.

Since mergemaster is included in the base system, I can directly commit
to its manpage.  There is already a description of `.mergemasterrc' in
the manpage, so I'm guessing you want a sample script added?

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

Re: congrlations to the freebsd developers

[Giorgos Keramidas]

On 2007-11-17 18:53, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
> M. Warner Losh wrote:
> > You can run FreeBSD/i386 on amd64 boxes.
>
> All the different methods people have suggested to me for doing this
> has blown up also since I use a nvidia card it is almost pointless
> unless I can run the nvidia kernel module.

That's odd.

The laptop I'm typing this is amd64-capable, but installing from the
FreeBSD/i386 release CD-ROMs worked like a charm.  Which methods have
you tried?

BTW, the i386 kernel module of nvidia should run fine on FreeBSD/i386.

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

Re: What do I put in fstab to get my DVD/CDROM burner to work?

[Giorgos Keramidas]

On 2007-11-17 14:13, Gary Kline <[EMAIL PROTECTED]> wrote:
>On Sat, Nov 17, 2007 at 08:03:25PM +0200, Giorgos Keramidas wrote:
>>On 2007-11-17 02:55, Joshua Isom <[EMAIL PROTECTED]> wrote:
>>> On Nov 16, 2007, at 10:56 PM, Yeef wrote:
>>>> this is work for me  freebsd 6.2-RELEASE
>>>>
>>>> /dev/acd0   /cdrom  cd9660  ro,noauto   0   0
>>>>
>>>> you should use root mount it.
>>>
>>> Or set vfs.usermount to 1, if I remember right.  I can't recall what's
>>> the proper method for setting it at boot, rc.conf or loader.conf.  The
>>> default is 0, which is what I have it set to, more to annoy me than
>>> security (personal server behind a buggy router/firewall).
>>
>>  man sysctl.conf
>>
>> That's the proper place to put `vfs.usermount=1'.
>
>  Okay, I've set vfs.usermount=1, but both totem and kmplayer
>  refuse to play my audio-CD.  Using #mount alone (as root)
>  doesn't say anything about /dev/acd0.  I have tried to mount
>  the CD ::
>
>  [EMAIL PROTECTED]:/dev# mount_cd9660 /dev/acd0 /media/cdroms/0
>  mount_cd9660: /dev/acd0: Invalid argument

You are not trying to _mount_ an audio CD-ROM, right?

If you are, then please read carefully the Handbook chapter about
multimedia and CD-ROM/DVD-ROM disks.  It will help a lot :)

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

Re: Making mergemaster skip certain files

[Giorgos Keramidas]

On 2007-11-16 22:34, "J. Porter Clark" <[EMAIL PROTECTED]> wrote:
> Is there any way to keep certain files out of the reach of
> mergemaster?  I understand the need for carefully merging the
> old and the new, but I really shouldn't ever have to for files
> like these:
>
>   /etc/aliases
>   /etc/hosts
>   /etc/hosts.allow
>   /etc/manpath.config
>   ... and many others.

You can hook into mergemaster, using shell scripts.  See the manpage of
mergemaster for more details, but here's a short description of what I
use on my laptop for some time now:

1. A `.mergemasterrc' file in the HOME directory of the `root' user,
   which contains:

STRICT=no
MM_PRE_COMPARE_SCRIPT=/root/mm-pre-compare.sh

2. The `/root/mm-pre-compare.sh' script contains the following:

#!/bin/sh

# NOTE: No PATH needed, because mm's PATH is already
# draconian enough.

# If TEMPROOT is not set, or it is set to a path which
# resolves to the real root filesystem, abort early, before we
# trash the config files of the installed root filesystem.

if test -z "${TEMPROOT}" ; then
echo >&2 "$0: error: TEMPROOT is unset or empty."
exit 1
fi
p=`realpath "${TEMPROOT}"`
if test "${p}" = '/' ; then
echo >&2 "$0: error: TEMPROOT is the real root filesystem."
exit 2
fi

case "${PRE_WORLD}" in
'')
# The following files always have local changes.
# Remove them from ${TEMPROOT} to force mergemaster(8)
# to ignore these files when comparing /etc directories.

rm -f "${TEMPROOT}/.cshrc"
rm -f "${TEMPROOT}/.profile"
rm -f "${TEMPROOT}/root/.cshrc"
rm -f "${TEMPROOT}/root/.profile"

rm -f "${TEMPROOT}/etc/hosts"
rm -f "${TEMPROOT}/etc/networks"

rm -f "${TEMPROOT}/etc/motd"
rm -f "${TEMPROOT}/etc/printcap"

;;
esac

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

Re: What do I put in fstab to get my DVD/CDROM burner to work?

[Giorgos Keramidas]

On 2007-11-16 22:24, Chris Hill <[EMAIL PROTECTED]> wrote:
>On Fri, 16 Nov 2007, David J Brooks wrote:
>>On Friday 16 November 2007 08:23:21 pm Gary Kline wrote:
>>> This is the old and current fstable:
>>>
>>> # DVD drive (top)
>>> /dev/acd0   /media/cdroms/0 udf ro,noauto   0   0
>>> # CD-burner (bottom)
>>> /dev/acd1   /media/cdroms/1 cd9660  ro,noauto   0   0
>>
>> cd9660 is what you need. for the burner at least though, you'll want
>> to make it rw rather than ro.

Not really.  When *mounted* even DVD-RW disks are read-only.

> Good point! Although my CD burner burns CDs just fine with either
> cdrecord or burncd, even with ro in its fstab line.

That's because they are not written ``through the cd9660 filesystem
driver'', but through cdrecord/burncd.

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

Re: What do I put in fstab to get my DVD/CDROM burner to work?

[Giorgos Keramidas]

On 2007-11-17 02:55, Joshua Isom <[EMAIL PROTECTED]> wrote:
> On Nov 16, 2007, at 10:56 PM, Yeef wrote:
>> this is work for me  freebsd 6.2-RELEASE
>>
>> /dev/acd0   /cdrom  cd9660  ro,noauto   0   0
>>
>> you should use root mount it.
>
> Or set vfs.usermount to 1, if I remember right.  I can't recall what's
> the proper method for setting it at boot, rc.conf or loader.conf.  The
> default is 0, which is what I have it set to, more to annoy me than
> security (personal server behind a buggy router/firewall).

man sysctl.conf

That's the proper place to put `vfs.usermount=1'.

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

Re: bash and strings

[Giorgos Keramidas]

On 2007-11-16 03:43, [EMAIL PROTECTED] wrote:
> Everyone,
>
> I'm sure this is easy, and I am making it harder than it is.
>
> I am being supplied a list of files, and need to create the files and
> directories to hold them, but I cannot figure out how to take the string
> apart.
>
> For example, I am given
>
> /usr/local/scripts/firewall.sh
>
> I need to create the /usr/local/scripts directory and then create
> firewall.sh.

See the `dirname' and `basename' commands:

$ dirname "/usr/local/scripts/firewall.sh"
/usr/local/scripts
$

$ basename "/usr/local/scripts/firewall.sh"
firewall.sh
$

Be careful about properly quoting the filenames though (note how the
first invocation of `dirname' fails below, and try to understand why
it fails):

$ testname='foo bar baz'
$ dirname $testname
usage: dirname string
$ dirname "${testname}"
.
$

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

Re: semi-OT: awk - field separator

[Giorgos Keramidas]

On 2007-11-15 13:24, Robert Huff <[EMAIL PROTECTED]> wrote:
> What is the canonical way to get (FreeBSD default) awk to use a
> single double-quote as the field separator?  I have tried variations
> on -F\\\" and -F"\"" and the best I can get is:
>
> + awk -F" {print $2}
> ./script.sh: 1: Syntax error: word unexpected

Any of the following should work:

awk -F\" ...
awk -F"\"" ...
awk -F'"' ...

I see that this is part of a script (the '+' character at the start of
the line).  Can you show us the *exact* contents of the script at the
point where awk(1) is called?

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

Re: OT: gcc/binutils question

[Giorgos Keramidas]

On 2007-11-15 17:17, Wojciech Puchar <[EMAIL PROTECTED]> wrote:
> how to convert arbitrary data (in file) to object file, so i will be able
> to do
>
> extern char something[]
>
> and use it - in C.

Try to file2c(1) utility.  Quoting from its manpage:

% EXAMPLES
%  The command:
%
%date | file2c 'const char date[] = {' ',0};'
%
%  will produce:
%
%const char date[] = {
%83,97,116,32,74,97,110,32,50,56,32,49,54,58,50,56,58,48,53,
%32,80,83,84,32,49,57,57,53,10
%,0};

> i did wrote converter that converts data from file to
>
> const char something[]={firstbyte,secondbyte,.};
>
> and then cc to compile it.
>
> and it's VERY SLOW when data are few megs - while it's no work in fact.

I think there's something wrong with the converter, then.  The file2c(1)
utility converts my kernel (6 MB of data) in less than 3 seconds:

$ /usr/bin/time file2c 'const char kernel[] = {' '};' \
< /boot/kernel/kernel > /dev/null
2.98 real 2.92 user 0.11 sys
$

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

Re: cron/send mail question

[Giorgos Keramidas]

On 2007-11-15 13:47, zbigniew szalbot <[EMAIL PROTECTED]> wrote:
> Dear all,
> What command (when using cron) should I invoke to automatically sent
> /var/log/exim/rejectlog file to a specified email address? I just need
> to analyze it and would best prefer to have it in my inbox in the
> morning.

There are several ways.

(1) Add a new shell script to `/usr/local/etc/periodic/daily'.

(2) Add the commands you want to run in `/etc/daily.local'.

(3) Add a cronjob in `/etc/crontab' to run a custom script.


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

Re: Socket programming question

[Giorgos Keramidas]

On 2007-11-14 15:21, Andrew Falanga <[EMAIL PROTECTED]> wrote:
> Hi,
> My question has to do with how someone would find out if a call to
> socket(2) actually produced a socket.  I know that the API works, I've
> programmed with it many times, but is there a way to find out if 's'
> returned by socket(2) is actually valid in whatever kernel structure
> it is stored?  I understand that I may have the process entirely mixed
> up.  But it seems to me that the socket is somehow known to the kernel
> and I should be able to query the kernel somehow and discover if it is
> valid.

The socket() system call returns -1 when it fails.  Isn't that
sufficient?  If not, why?  What ``extra'' information would you
expect from the kernel when socket() fails?

> Let me know if my question doesn't make sense as worded and I'll try
> to explain myself better.  Another question related to this one, would
> someone in this list know where the source code is, in the system
> source tree, for the select call?

Look for kern_select() in `/usr/src/sys/kern/sys_generic.c'.

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

Re: Gcc Problems

[Giorgos Keramidas]

On 2007-11-14 13:21, Mike Fahey <[EMAIL PROTECTED]> wrote:
>
> Freebsd4.10
>
> When I try to compile any problem I get the following error. Any
> suggestions? Thanks.
>
> helpers/dummy.c:1: syntax error before `/'
>
> Here is a sample trying to build any port.
>
> make
> ===>   wget-1.10.2_1 depends on file: /usr/local/lib/libcrypto.so.4 -
> found
> ===>   wget-1.10.2_1 depends on executable: gmake - found
> ===>   wget-1.10.2_1 depends on file: /usr/local/bin/perl5.8.8 - found
> ===>   wget-1.10.2_1 depends on shared library: intl - found
> ===>  Configuring for wget-1.10.2_1
> configure: WARNING: you should use --build, --host, --target
> configure: configuring for GNU Wget 1.10.2
> checking build system type... i386-portbld-freebsd4.10
> checking host system type... i386-portbld-freebsd4.10
> checking whether gmake sets $(MAKE)... eval: 1: Syntax error: "("
> unexpected
> ===>  Script "configure" failed unexpectedly.
> Please report the problem to [EMAIL PROTECTED] [maintainer] and attach the
> "/usr/ports/ftp/wget/work/wget-1.10.2/config.log" including the output
> of the failure of your make command. Also, it might be a good idea to
> provide an overview of all packages installed on your system (e.g. an
> `ls /var/db/pkg`).
> *** Error code 1

The error message points to files which you can skim through, to see
more error messages.  Did you look at these?  If yes, can you attach a
copy and send it to this list and the maintainer email address listed
above?

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

Re: What server for a mail server

[Giorgos Keramidas]

On 2007-11-14 09:04, Olivier Nicole <[EMAIL PROTECTED]> wrote:
> Hi,
> I have a project where I should set-up a mail server for approximately
> 12000 users, 4000 virtual domains, 15000 messages per day, 700 MB
> traffic per day.
>
> The machine should run Postfix, courrier-imap and a web mail (probably
> squirel), Apache 2, spamassassin, clamav, greylist
>
> What type of machine should I target, in terms of CPU and RAM?

If you saturate an 100 Mbit/s network interface, it can push/pull more
than 800 GB per day, so it's not the network that is going to be the
problem :)

Depending on the number of *concurrent* connections you want to support,
and bearing in mind the post-processing you want to do (filtering with
clamav, spamassassin, and serving web pages through the web interface),
it may be worth aiming for a machine which has SMP support, and a fair
amount of physical memory.

I'd go for at least a 3 GHz processor, and 4 GB of physical memory.
If amd64 is an option, you can even shoot for 8 GB of RAM or more.

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

Re: problems with building a patch

[Giorgos Keramidas]

On 2007-11-11 16:02, Matthias Apitz <[EMAIL PROTECTED]> wrote:
>
> Hello,
>
> I've build a patch for 'nn-6.7.3' to add support for RFC1522 to my
> beloved news-reader. Before giving it away I was trying it on a fresh
> workspace of the /usr/ports/news/nn and run into the problem that
> new files which brings the patch to the tree are always created
> in the current working dir, even if I create them before with touch(1),
> existing files, like 'answer.c' in the example below, get patched
> correctly:
>
> $ /usr/ports/news/nn/work
> $ touch nn-6.7.3/PATCH.RFC1522
> $ patch < ../myRFC1522.patch
> Hmm...  Looks like a unified diff to me...
> The text leading up to this was:
> --
> |diff -N -r -u -X exclude nn-6.7.3/PATCH.RFC1522 
> nn-6.7.3.patched/PATCH.RFC1522
> |--- nn-6.7.3/PATCH.RFC1522 Thu Jan  1 01:00:00 1970
> |+++ nn-6.7.3.patched/PATCH.RFC1522 Sat Nov 10 11:04:58 2007
> --

Here's the problem.

The patch files for ports should *not* include the `nn-6.7.3' part, like
this one.  They should be relative to the toplevel directory of the
unzipped/untarred port, i.e.:

diff -N -u PATCH.RFC1522.orig PATCH.RFC1522
--- PATCH.RFC1522.orig Thu Jan  1 01:00:00 1970
+++ PATCH.RFC1522 Sat Nov 10 11:04:58 2007

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

Re: need an explanation of "make serarch" output

[Giorgos Keramidas]

On 2007-11-11 09:23, Aryeh Friedman <[EMAIL PROTECTED]> wrote:
> Sorry beinf rude but I wonder if either of the responders took there
> own advice and RTFM'ed (where M=mail)
>
> What I mean is I was *NOT* asking how to get the deps list... I was
> asking that for example "make search key=jdk16 display=bdeps" will
> display multiple "bdeps" lines which one contains the correct deps
> list (if any) or do I need to combine them to get a complete list?

Yes, you need to combine them.

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

Re: nanobsd, picobsd, tinybsd

[Giorgos Keramidas]

On 2007-11-10 15:45, John Smith <[EMAIL PROTECTED]> wrote:
>On Nov 10, 2007 9:40 AM, Giorgos Keramidas <[EMAIL PROTECTED]> wrote:
>>On 2007-11-09 17:01, John Smith <[EMAIL PROTECTED]> wrote:
>>> Can anybody please explain to my what the differences are between
>>> nanobsd, picobsd and tinybsd.
>>>
>>> They all seem to be doing the same (creating a minimal FreeBSD image
>>> that can be used in embedded systems), or is this not right?
>>
>> What don't you experiment with them, and see? :)
>
> I'd expected a more level headed reply from this FreeBSD list. How is
> a newbie supposed to know the differenced and how can I test this if I
> don't have a spare machine?
>
> My question was more out of interest. This mailing list is called
> FreeBSD-Questions, so why can't I asked a reasonable question and
> expect a reasonable reply...?

Minus the typo in the original reply [s/What/Why/], I'm afraid that this
is the only way to get hands-on experience with these systems.  I'm
sorry if the original response came out as ``odd''.  More over, I don't
really know what you mean by ``level headed'', other than ``not flame me
for asking'', which is not something we tend to do in freebsd-questions.

Having said that, a brief description of what each one of the systems
you mentioned is:

  * PicoBSD used to work with earlier FreeBSD versions.  I don't think
it does work with recent 7.X versions or CURRENT.  I wouldn't even
go there right now, unless you want to ``forward port'' all the code
which made PicoBSD tick, and make it work with recent FreeBSD
releases.

  * NanoBSD is more ``modern'' than PicoBSD, and it works with 6.X, 7.X
and CURRENT releases.  One of the advantages of NanoBSD is that it's
part of the base system and it is easy to use.  Our documentation
includes an article about NanoBSD at:
<http://www.freebsd.org/doc/en_US.ISO8859-1/articles/nanobsd/>
Try reading the introduction of the article for some of the features
of NanoBSD.

  * TinyBSD is a set of tools and scripts, which is also part of the
base system.  It definitely has a few good characteristics.  For
instance their documentation is Wiki-like and gets updated often.

These short descriptions contain stuff that only scratches the surface
of what it *feels* like to work with each system.  If that's what you
originally wanted, then Google and ten minutes or so would do fine.  The
important bits, however, are always in the details, and that's why you
have to try NanoBSD and TinyBSD yourself, and see how much you like or
dislike each one of them.

- Giorgos

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

Re: nanobsd, picobsd, tinybsd

[Giorgos Keramidas]

On 2007-11-09 17:01, John Smith <[EMAIL PROTECTED]> wrote:
> Can anybody please explain to my what the differences are between
> nanobsd, picobsd and tinybsd.
>
> They all seem to be doing the same (creating a minimal FreeBSD image
> that can be used in embedded systems), or is this not right?

What don't you experiment with them, and see? :)

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

Re: Dangers of using a non-base shell

[Giorgos Keramidas]

On 2007-11-09 18:10, Alex Zbyslaw <[EMAIL PROTECTED]> wrote:
> Giorgos Keramidas wrote:
>> i.e. here's an ftp session on my laptop:
>>
>>  [EMAIL PROTECTED]:/root# fgrep ftp: /etc/passwd
>>  ftp:*:1003:1003:& user:/home/ftp:/usr/sbin/nologin
>>  [EMAIL PROTECTED]:/root# su ftp
>>  [EMAIL PROTECTED]:/root$ id
>>  uid=1003(ftp) gid=1003 groups=1003
>>  [EMAIL PROTECTED]:/root$
>
> Must be new, because in 5.4 I get:
> [...]
> I find the behaviour you get definitely undesirable.  There are
> occasionally accounts have special purpose shells which do work in
> some restricted fashion which you *might* want to use (in which case
> you can su) or which you might not (so you su -m). [...]

False alarm.  I had a desynced /etc/pwd.db when this happened.

The correct behavior with nologin as the shell is:

  [EMAIL PROTECTED]:/root# su ftp
  This account is currently not available.
  [EMAIL PROTECTED]:/root#

> Confused.

I apologize for the confusion :/


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

Re: Dangers of using a non-base shell

[Giorgos Keramidas]

On 2007-11-09 16:34, Alex Zbyslaw <[EMAIL PROTECTED]> wrote:
>
> [ discussing `su -m' option ]
>
> Also the only way I know on FreeBSD to interactively become a user
> with no real shell (true, nologin etc).

It should be possible to type:

su username

i.e. here's an ftp session on my laptop:

[EMAIL PROTECTED]:/root# fgrep ftp: /etc/passwd
ftp:*:1003:1003:& user:/home/ftp:/usr/sbin/nologin
[EMAIL PROTECTED]:/root# su ftp
[EMAIL PROTECTED]:/root$ id
uid=1003(ftp) gid=1003 groups=1003
[EMAIL PROTECTED]:/root$

Good idea, though :)

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

Re: Dangers of using a non-base shell

[Giorgos Keramidas]

On 2007-11-09 18:55, Andrew Pantyukhin <[EMAIL PROTECTED]> wrote:
> On Tue, Oct 30, 2007 at 01:39:12PM +0200, Giorgos Keramidas wrote:
> > I've been using the following for some time:
> >
> > keramida> su -
> > Password: 
> > root# exec env SHELL=/usr/local/bin/bash bash -l
>
> I know it doesn't work on slolaris^W some Unix flavors, but I've
> been quite happy with "su -m".

Heh, putting the Solaris bashing (sic) aside, I can see how the -m
option can be useful some times.  After all, it was implemented because
*someone* thought it would be neat to have around :-)


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

Re: How can I use the Windows loader to boot FreeBSD

[Giorgos Keramidas]

On 2007-11-07 15:19, Alou Dialy <[EMAIL PROTECTED]> wrote:
> I read this FAQ but I am still confused. How does it
> work if you have windows on ad0 and freebsd on ad1.
>
> http://www.freebsd.org/doc/en_US.ISO8859-1/books/faq/disks.html#NT-BOOTLOADER

AFAICT, the 'explicit' list of steps missing from the FAQ is:

  1. Run boot0cfg from FreeBSD:

# boot0cfg -B /dev/ad1

  2. Copy the MBR of the ad1 disk to a file:

# dd if=/dev/ad1 of=bootsect.bsd bs=512 count=1

  3. Save this file to a floppy disk, or some other Windows-accessible
 place.  Using a floppy-disk, formatted as FAT, this could be done
 by typing:

# mount -t msdosfs /dev/fd0 /mnt/dos
# cp bootsect.bsd /mnt/dos
# umount /mnt/dos

  4. Boot into Windows and copy the file A:\BOOTSECT.BSD to C:\

  5. Continue as described in the FAQ :)


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

Re: New FreeBSD art?

[Giorgos Keramidas]

On 2007-11-07 13:49, James <[EMAIL PROTECTED]> wrote:
> I know for sure. We shall rue the day beastie was ever demoted
> to mere mascot.
>
> RUE I TELL YOU

The beastie is here[1] to stay, no worries...
[1] 'Here' as in anywhere I can attach a sticker :-)

___
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 write a condition in Bourne shell

[Giorgos Keramidas]

On 2007-11-05 14:03, Olivier Nicole <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I am a lame Bourne sheel programmer, how to write:
>
> while [ ( $? -ne 0 ) -a ( $retry -gt 0 ) ] ; do
>
> that should execute as long as $? is not null and $retry is greater
> than 0?

Try something like...

retry=0
done=-0
while [ $done -eq 0 ] && [ $retry -lt 10 ]; do

run_some_other_stuff_here

if [ $? -eq 0 ]; then
done=1
fi
done

if [ $done -eq 0 ]; then
echo "Failed after $retry attempts."
fi

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

Re: Is there a way to compare what is in the ports tree with what is installed?

[Giorgos Keramidas]

On 2007-11-05 02:36, Pollywog <[EMAIL PROTECTED]> wrote:
>>  dir /var/d/pkg | grep 
>
> My Linux systems have a "dir" command but my FreeBSD does not.
> Is there something I need to install?

Not really.  The ls(1) utility works fine :-)

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

Re: Is there a way to compare what is in the ports tree with what is installed?

[Giorgos Keramidas]

On 2007-11-05 14:53, Brett Davidson <[EMAIL PROTECTED]> wrote:
> ie. If I had a particular version of the ports tree on a server, how
> could I check to see if any of the programs in that tree were actually
> installed?
>
> Is there a simple command or sequence of commands to do this?

Try running the pkg_version(1) utility.  For instance, running it here,
I can see output like this:

ksh$ pkg_version | grep -v '=$'
ORBit2  <
aalib   <
[...]

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

Re: reverse grep

[Giorgos Keramidas]

On 2007-11-04 01:12, [EMAIL PROTECTED] wrote:
> How is it possible to select lines that do NOT match a specific pattern?

grep -v 'pattern'

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

Re: memory pool, rfc

[Giorgos Keramidas]

On 2007-11-01 02:06, Eduardo Morras <[EMAIL PROTECTED]> wrote:
> At 01:38 01/11/2007, you wrote:
>> This is a 'general' questions list, with a fairly high level of traffic.
>> Many FreeBSD committers and team affiliates are subscribed, but there
>> are still a lot of knowledgeable people who are respected team members,
>> but are *not* subcribed here.
>>
>> Maybe in a more appropriate list you can get a greater amount of
>> visibility with people who can help your contributions to be integrated
>> with the main source code tree of FreeBSD.
>
> Understood
>
>> > I think that in this list are FreeBSD *gurus*/hacks too which could
>> > say a "try it" or a "are you crazy?" answer. If other developers
>> > thinks that they need my rfc i'll add my code to FreeBSD.
>>
>> It's hard to tell in advance, without actually *seeing* something in the
>> form of a proposal, or at least an implementation.
>> .
>> testing process of the source integration, and so on.
>>
>> This means that the next reasonable step would be to make available at
>> least a proposal or source patch, and post it to one of the developer
>> lists of FreeBSD.  The freebsd-hackers list would be nice.
>
> I can prepare a malloc-like pilot implementation and detailed paper.

That would be fantastic!

>> It doesn't sound like a productive use of everyone's time, unless we
>> have a solid plan for a better zlib replacement in place :-)
>
> You can check http://www.maximumcompression.com/index.html and
> http://cs.fit.edu/~mmahoney/compression/text.html for the state of the
> art lossless compressors.

I will, thanks for the references; they look interesting :)



pgp4BbuIrO69j.pgp
Description: PGP signature

Re: memory pool, rfc

[Giorgos Keramidas]

On 2007-11-01 01:00, Eduardo Morras <[EMAIL PROTECTED]> wrote:
> I'm already on hackers list but has very low traffic (9 messages last
> 5 days) and this is a question list no ? ;-)

This is a 'general' questions list, with a fairly high level of traffic.
Many FreeBSD committers and team affiliates are subscribed, but there
are still a lot of knowledgeable people who are respected team members,
but are *not* subcribed here.

Maybe in a more appropriate list you can get a greater amount of
visibility with people who can help your contributions to be integrated
with the main source code tree of FreeBSD.

> I think that in this list are FreeBSD *gurus*/hacks too which could
> say a "try it" or a "are you crazy?" answer. If other developers
> thinks that they need my rfc i'll add my code to FreeBSD.

It's hard to tell in advance, without actually *seeing* something in the
form of a proposal, or at least an implementation.

The FreeBSD team works in an open manner, and we take pride in the
fairly well established process of development we use.  Everyone is
invited to support, extend, fix bugs, implement new features, or
otherwise improve FreeBSD, as long as they are prepared to put in some
effort to "fit within the team".  Having said that, all you would have
to show us would be at least one of:

* A detailed proposal of the changes, optionally including a
  justifiction of why you think the changes are necessary and
  how they will improve the user experience of the people who
  comprise our userbase.

* A pilot implementation of the new features you described.
  It doesn't have to be complete; it doesn't have to precisely
  match the existing style of the code; it doesn't have to be
  performing blazingly fast, if there are plans for tuning;
  it doesn't even have to be compilable; as long as you are
  prepared to help us fix all the details.

Then, I'm sure that if the improvements you are working on look
interesting, you stand a better chance of soliciting comments, feedback
related to the way the changes can go into FreeBSD, help with the
testing process of the source integration, and so on.

This means that the next reasonable step would be to make available at
least a proposal or source patch, and post it to one of the developer
lists of FreeBSD.  The freebsd-hackers list would be nice.

> The point to zlib / libbzip2 is because when i talk about compression
> *always* someone says " Why don't use zlib?" Short answer No, Medium
> Answer: deflate is 15 years old and there are better and faster algs
> now.

That's one way to describe zlib.  Perhaps not the only one, but still a
valid opinion to hold, so let's go down that path.  It doesn't sound
like a productive use of everyone's time, unless we have a solid plan
for a better zlib replacement in place :-)

Giorgos



pgpHhBKCvTiOW.pgp
Description: PGP signature

Re: Core dump

[Giorgos Keramidas]

On 2007-10-31 15:32, Rem P Roberti <[EMAIL PROTECTED]> wrote:
> A program that I use has started giving me this error message when I try
> to load it:
>
> Segmentation fault: 11 (core dumped)
>
> Can someone give me a heads up on what's going on here.  I've done a
> reinstall to no avail.

Is there any reason why you can't tell us the name of the program?

The more details you provide, the easier it is going to be for someone
else to help you troubleshoot the problem.

Do you have a debug-build of the program around?  If you do, you can use
gdb(1) to try to grab a stack trace of the program, and with a little
luck and a bit of effort, it may even be possible to find out the root
cause of the segmentation fault.

Giorgos

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

Re: release(8) environmental variables

[Giorgos Keramidas]

On 2007-10-31 13:26, "Brian A. Seklecki" <[EMAIL PROTECTED]> wrote:
>>> really sucks.
>> 
>> I believe that's intentional, so re-running "make release" with
>> different CHROOTDIR values will produce consistently "similar" binaries.
> 
> I use LOCAL_SCRIPT to copy /etc/make.conf (well, /etc/src.conf) into place 
> inside the jail for the rebuild so that I can build a custom internal 
> release w/o certain subsystems (IPv6 or CSH, for example)
> 
> Is there a better way to do it?

LOCAL_SCRIPT sounds fine to me :)

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

Re: release(8) environmental variables

[Giorgos Keramidas]

On 2007-10-31 11:03, "Brian A. Seklecki" <[EMAIL PROTECTED]> wrote:
>On Wed, 2007-10-31 at 13:54 +0200, Giorgos Keramidas wrote:
>>On 2007-10-30 18:02, "Brian A. Seklecki" <[EMAIL PROTECTED]> wrote:
>>>
>>> As far as building goes, the variables in play are:
>>>
>>>   DESTDIR, MAKEOBJDIRPREFIX, RELEASEDIR, CHROOTDIR, CVSROOT,
>>>   EXTSRCDIR, EXTDOCDIR, BUIILDNAME, RELEASETAG, NODOC, NOPORTS,
>>>   WORLD_FLAGS, LOCAL_SCRIPT
>>>
>>> For stage one of the release process, the following seem relevant:
>>>
>>>   DESTDIR, MAKEOBJDIRPREFIX, RELEASEDIR
>>>
>>> For stage two:
>>>
>>>   CHROOTDIR, CVSROOT, EXTSRCDIR, EXTDOCDIR, BUIILDNAME, RELEASETAG, NODOC,
>>>   NOPORTS, WORLD_FLAGS, LOCAL_SCRIPT
>>>
>>> Do you guys prefer to set these in make.conf(5) or as exported
>>> environmental variables in the shell that spawn's make(1) ?
>>
>> make.conf is too invasive.  I just set them in a shell script called
>> `bldenv.sh' and saved in the release-checkout area :)
> 
> I ask because I noticed that the following variablkes do not get
> honored by "make release" that occurs inside the chroot() as spawned
> by "make release" (presumably during release.5)
> 
> DESTDIR, MAKEOBJDIRPREFIX, RELEASEDIR 
> 
> For example, OBJs get sent right into $CHROOTDIR/usr/obj/, which
> really sucks.

I believe that's intentional, so re-running "make release" with
different CHROOTDIR values will produce consistently "similar" binaries.


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

Re: release(8) environmental variables

[Giorgos Keramidas]

On 2007-10-30 18:02, "Brian A. Seklecki" <[EMAIL PROTECTED]> wrote:
>
> As far as building goes, the variables in play are:
>
>   DESTDIR, MAKEOBJDIRPREFIX, RELEASEDIR, CHROOTDIR, CVSROOT,
>   EXTSRCDIR, EXTDOCDIR, BUIILDNAME, RELEASETAG, NODOC, NOPORTS,
>   WORLD_FLAGS, LOCAL_SCRIPT
>
> For stage one of the release process, the following seem relevant:
>
>   DESTDIR, MAKEOBJDIRPREFIX, RELEASEDIR
>
> For stage two:
>
>   CHROOTDIR, CVSROOT, EXTSRCDIR, EXTDOCDIR, BUIILDNAME, RELEASETAG, NODOC,
>   NOPORTS, WORLD_FLAGS, LOCAL_SCRIPT
>
> Do you guys prefer to set these in make.conf(5) or as exported
> environmental variables in the shell that spawn's make(1) ?

make.conf is too invasive.  I just set them in a shell script called
`bldenv.sh' and saved in the release-checkout area :)

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

Re: [freebsd-questions] Dangers of using a non-base shell

[Giorgos Keramidas]

On 2007-10-30 20:39, Howard Jones <[EMAIL PROTECTED]> wrote:
> Benjamin M. A'Lee wrote:
>> You could possibly also put "bash -l && exit" in your .shrc, which would
>> exit if bash exited successfully. I haven't tested it, but it should
>> work.
>
> or 'exec bash -l' which will replace the existing shell with bash in
> memory, rather than run it from it as a subprocess. I was going to verify
> that that's the technical explanation, but 'man exec' gets you the utterly
> useless builtin(1) manpage.

It is a fairly ok description of the technical behavior.  See my similar
suggestion for using:

tcsh# exec env SHELL=/usr/local/bin/bash bash -l
bash#

> The effect is that you only have to type exit once, anyway.

Yup :)

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

Re: Dangers of using a non-base shell

[Giorgos Keramidas]

On 2007-10-29 20:50, Stephen Allen <[EMAIL PROTECTED]> wrote:
> It's been drawn to my attention not to use bash from the ports
> collection, because if one of it's dependencies (gettext or libiconv)
> fails or is updated significantly, it could break, and prevent
> login. The suggested solution was to use a base shell (such as sh) and
> append 'bash -l' to .shrc to automatically enter bash.
>
> The quite annoying side-effect is having to type 'exit' twice to get
> out of a su shell or screen.
>
> Would it be a better idea to use the pre-compiled binary for bash?
> And if I did so, could I be alerted to updates as easy as using
> 'pkg_version -v' when checking if any ports need updating?

I've been using the following for some time:

keramida> su -
Password: 
root# exec env SHELL=/usr/local/bin/bash bash -l

The same trick works with s/bash/mksh/ or s/bash/pdksh/, as far as I can
tell, and tcsh's history mechanism makes it easy to run the same command
later.  All I have to type is "exec " and hit the arrow-up key :)

The env(1) trick makes sure that sub-shells use bash(1) too, and 'exec'
reduces the number of 'exit' commands I have to type.

Now, if there's a problem with bash(1), I will probably have to su
again, but that's less trouble than being locked out because gettext has
broken, the *default* root shell is bash and _that_ fails all the time.

- Giorgos

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

Re: evil idea

[Giorgos Keramidas]

On 2007-10-27 21:49, mv <[EMAIL PROTECTED]> wrote:
> On 10/26/07, Aryeh M. Friedman <[EMAIL PROTECTED]> wrote:
> > I am running amd64 8-CURRENT and there are a few i386 only ports that I
> > absolutely must have installed and at the same time since I have
> > 4 GB of RAM all kinds of bizarreness is created if I "downgrade"
> > to i386.   So here is the idea:   use qemu to create a virtual
> > version of my machine (with less then 2GB or RAM) and install
> > i386 8-CURRENT on it (I want to use -CURRENT for all my installs)
>
> On Sat, 27 October 2007 20:33:44 Giorgos Keramidas wrote:
> > I don't think Aryeh wants to "install i386 over his current amd64".
> >
> > What he seems to be asking is if he can run *two* versions, one of
> > them as a virtualized host under qemu.  That should work, AFAIK.
>
> Would installing Freebsd i386 within a jail on an amd64 host solve his
> problem?

I don't know if that would 'solve' the problem, but one major difference
from qemu is that a jailed host runs the same kernel with the master.

This means that i386 programs do not run in 'native' i386 mode, but in
emulated mode.  The FreeBSD kernel already includes support for running
i386 programs in emulated mode, but since I haven't done this I am not
sure how feasible or useful it is.

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

Re: evil idea

[Giorgos Keramidas]

On 2007-10-26 23:34, Astrodog <[EMAIL PROTECTED]> wrote:
>On 10/26/07, Aryeh M. Friedman <[EMAIL PROTECTED]> wrote:
>> I am running amd64 8-CURRENT and there are a few i386 only ports that I
>> absolutely must have installed and at the same time since I have 4 GB of
>> RAM all kinds of bizarreness is created if I "downgrade" to i386.   So
>> here is the idea:   use qemu to create a virtual version of my machine
>> (with less then 2GB or RAM) and install i386 8-CURRENT on it (I want to
>> use -CURRENT for all my installs)
>>
>> Any thing I should watch out for here (I know I need to use NFS or
>> something like it to share files between the host and guest OS's)
> 
> There's actually a known system that will work for this. You can use
> your existing swap partition, as an "extra" root partition, installing
> there, then booting to that, then rebuild/install to your original
> partition. Its the same basic idea as the method for updating from
> 4.x->7.x, and should be on the lists. (Note to docs, might be worth
> putting it somewhere.)

I don't think Aryeh wants to "install i386 over his current amd64".

What he seems to be asking is if he can run *two* versions, one of
them as a virtualized host under qemu.  That should work, AFAIK.

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

Re: Mentor for C self study wanted

[Giorgos Keramidas]

On 2007-10-23 23:24, Harald Schmalzbauer <[EMAIL PROTECTED]> wrote:
> Thanks all,
> here was my example, just for completeness, I found mentors for my
> needs.

> #include 
>
> void main()
> {
>   short nnote;
>
>   // Numerischen Notenwert einlesen
>   printf("Bitte numerischen Schulnotenwert eingeben: ");
>   scanf("%d",&nnote);

You are passing "%d" to scanf() so it expects to find enough 'storage'
in its pointer argument for an 'int'.  If 'short' happens to have a
smaller size (as is commonly the case), scanf() will overwrite random
memory locations after 'nnote'.  On systems where 'nnote' is stored in
the stack (because it's an automatic/local variable of main()), you are
risking stack corruption (and a SEGFAULT *may* happen).

It's also a very good idea to check the return code of scanf():

int nnote;

if (scanf("%d", &nnote) != 1) {
error;
}

>   switch (nnote)
>   {
> case 1: printf("Die Note %d entspricht sehr gut.",nnote);
> break;
> case 2: printf("Die Note %d entspricht gut.",nnote);
> break;
> case 3: printf("Die Note %d entspricht befriedigend.",nnote);
> break;
> case 4: printf("Die Note %d entspricht ausreichend.",nnote);
> break;
> case 5: printf("Die Note %d entspricht mangelhaft.",nnote);
> break;
> case 6: printf("Die Note %d entspricht ungen?gend.",nnote);
> break;
> default: printf("%d ist keine zul?ssige Schulnote!");

There's no `int' argument to the printf() call of the default clause.

This will either cause printf() to print random garbage, or try to
access memory regions which are unmapped and SEGFAULT.

> P.S.:
> I found that declaring nnote as int soleves my problem, but I couldn?t
> understand why.
> Another one was the result of default: nnote was -1077942208 instead
> of 9 for example.

It was never assigned to 9 :)

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

Re: Q: general LaTeX mailing list

[Giorgos Keramidas]

On 2007-10-23 12:43, Byung-Hee HWANG <[EMAIL PROTECTED]> wrote:
> Hi there,
>
> Who knows a good general LaTeX mailing list? Ah yes, here is also good
> mailing list for the question. However, I want to give specific and
> professional advice about LaTeX. Unfortunately, Google disappointed my
> desire ;;

If you don't have a dislike for newsgroups, then ``news:comp.text.tex''
is a pretty good choice.

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

Re: Want to upgrade sendmail in next OS release

[Giorgos Keramidas]

On 2007-10-21 16:42, "Tuc at T-B-O-H.NET" <[EMAIL PROTECTED]> wrote:
>>On 2007-10-21 11:28, "Tuc at T-B-O-H.NET" <[EMAIL PROTECTED]> wrote:
>>> Hi,
>>>
>>> I'd like to upgrade my sendmail version in advance of upgrading to the
>>> next release of the OS.
>>
>> Why?
>
> I can afford the time to start/stop sendmail here and there, but I
> have to schedule the maintenance window to upgrade this particular
> machine.

I see.  I'd still go for the maintenance window option, unless there is
a _very_ pressing need to upgrade Sendmail *today* because of a security
update, but you have a point :)

>>> # cd /usr/src/lib/libsm
>>> # make obj && make depend && make
>>> # cd /usr/src/lib/libsmutil
>>> # make obj && make depend && make
>>> # cd /usr/src/usr.sbin/sendmail
>>> # make obj && make depend && make && make install
>>
>> This can be done only if you manage to upgrade Sendmail sources and
>> their build-glue manually, by following *exactly* the same process that
>> Gregory Shapiro (our native Sendmail maintainer) uses to do it.
>
> So where do I find that? I took those instructions from previous
> FreeBSD Security announcements against sendmail.

The process is described in `/usr/src/contrib/sendmail/FREEBSD-upgrade',
but it takes a certain amount of experience with CVS and merging.  Even
if you don't really want to use CVS to merge the changes yourself, the
list of build/install/runtime infrastructure files near the end of this
file will prove veery useful.  Quoting from a recent copy:

% The following files make up the sendmail build/install/runtime
% infrastructure in FreeBSD:
%
% Makefile.inc1
% bin/Makefile
% bin/rmail/Makefile
% contrib/sendmail/
% [...]

>> You are also missing some parts of Sendmail above, like:
>>
>>  /usr/src/bin/rmail
>>  /usr/src/usr.bin/vacation
>>  /usr/src/etc/sendmail
>
>   rmail - 99% of delivery is procmail, but good point.
>   vacation - Not used
>   etc/sendmail - I don't use default mc's, and don't mind
>   running older cf's on a newer version. The
>   change between 8.13 and 8.14 wasn't enough for
>   us to need to.

Interesting bits of that list are:

lib/libmilter/Makefile
lib/libsm/Makefile
lib/libsmdb/Makefile
lib/libsmutil/Makefile
libexec/mail.local/Makefile
libexec/smrsh/Makefile
usr.bin/vacation/Makefile
usr.sbin/editmap/Makefile
usr.sbin/mailstats/Makefile
usr.sbin/makemap/Makefile
usr.sbin/praliases/Makefile
usr.sbin/sendmail/Makefile
usr.sbin/mailwrapper/Makefile

For future upgrades of Sendmail, it would probably be a good idea to
upgrade the libraries *first* and only when you are done building the
new libraries to install everything.

It may be possible to build everything with MAKEOBJDIRPREFIX='/usr/obj',
and install in one go when you are done with everything, but that's
something you should probably experiment a bit with -- preferrably in a
test machine, before you do the same on a live system.

> In the mean time, I got bored, so I did just that. Seems to be working
> fine, has processed about 15K emails since.

Neat :)

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

Re: Want to upgrade sendmail in next OS release

[Giorgos Keramidas]

On 2007-10-21 11:28, "Tuc at T-B-O-H.NET" <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I'd like to upgrade my sendmail version in advance of upgrading to the
> next release of the OS.

Why?

> # cd /usr/src/lib/libsm
> # make obj && make depend && make
> # cd /usr/src/lib/libsmutil
> # make obj && make depend && make
> # cd /usr/src/usr.sbin/sendmail
> # make obj && make depend && make && make install

This can be done only if you manage to upgrade Sendmail sources and
their build-glue manually, by following *exactly* the same process that
Gregory Shapiro (our native Sendmail maintainer) uses to do it.

You are also missing some parts of Sendmail above, like:

/usr/src/bin/rmail
/usr/src/usr.bin/vacation
/usr/src/etc/sendmail

The upgrade process used by Gregory Shapiro and the normal 'buildworld'
cycle takes care of all that and more.

Why would you want to circumvent it, and risk breaking Sendmail?

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

Re: sendmail problems

[Giorgos Keramidas]

On 2007-10-19 00:23, Steve Bertrand <[EMAIL PROTECTED]> wrote:
>Duane Winner wrote:
>> Hello,
>>
>> I need help to resolve a problem with my sendmail server.
>
> Usually when 'sendmail' is in a subject line, one can usually expect
> Giorgos respond :)

Thanks Steve, I'm honored by the confidence :)

I've been having network connectivity issues and fell a lot back in my
email backlog.  I'm back online, with a better setup now, so things will
start improving I guess.

> This is not a Sendmail issue, per-se.

You are right about that.  The original email by Duane Winner said:

I need help to resolve a problem with my sendmail server.

In my /var/log/maillog, I've been seeing:

sm-mta[1753]: l9H4EoAn001753: outbound-mail-10.bluehost.com
[69.89.17.210] did not issue MAIL/EXPN/VRFY/ETRN during connection
to IPv4

This usually means that there is an incoming connection from the host
outbound-mail-10.bluehost.com, whose IP address is 69.89.17.210, but the
connection was lost before the host managed to issue any commands to
deliver email, query for alias expansion, and so on.

Whenever I've seen this happening, it is usually some sort of network
setup error, broken routing, a misconfigured router in the path between
the originating host (outbound-mail-10.bluehost.com) and Sendmail, or
something similar.

> If we can take this back one step to re-evaluate the entire situation
> it may help:
>
> - what is the domain you are seeing issues with (are there more than
>   one? If so, are they on the same box/IP?)
> - what IP is this domain's mail operating on
> - examples of domains you see problems with, and examples of those you don't
> - is it only mailing lists you have problems with
> - do you receive this email I am sending on the problematic server
>
> Do you have another site that you can confirm working/not working?

Good points :)

Duane, can you respond to the questions of Steve above?  They will at
least provide us with hints to start troubleshooting this better.

- Giorgos

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

Re: Why 7.0 is so late ?

[Giorgos Keramidas]

On 2007-10-18 18:31, Robert Huff <[EMAIL PROTECTED]> wrote:
>Jerry McAllister writes:
>>> I'd much rather that a RELEASE version is as stable as it can
>>> reasonably be made than that it arrives "on time".
>>
>> Yup.  I think that is the way all of us feel.
>
> I don't think many will argue, at least not loudly.  However ... many
> also have bad memories of 5.0, and the grim firm desire to never
> let that happen again.

Indeed.  That's one of the most important driving forces between trying
to push major releases out with a period of "around 18 months" ;)

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

Re: Ref : Permission to use parts of FreeBSD documentation on My Website

[Giorgos Keramidas]

On 2007-10-17 17:31, "Aryeh M. Friedman" <[EMAIL PROTECTED]> wrote:
>allen paul wrote:
>> Thank you, That was the peice of information I was looking for.
>> Sure, I will respect proper credits, and skip copyrighted
>> information.  I might have been mistaken if I mistook BSD
>> documentation as freely distributed information under GNU license.
>> But, as long as I can post a few Kernel related excerpts, I think I
>> am good.
>
> In general your going to find BSD a more permissive license then GPL
> (for sure less viral).  Contact me privately if you need a good
> background references this as well other possible licensing options
> you may have.

... or ask here, on the freebsd-questions mailing list.  There is no
reason why a discussion about the way the BSD license works would be
kept `private' in personal mailboxes :)

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

Re: Ref : Permission to use parts of FreeBSD documentation on My Website

[Giorgos Keramidas]

On 2007-10-18 02:04, allen paul <[EMAIL PROTECTED]> wrote:
> Hi,
> I working on compiling Linux kernel and related issues, I was
> interested in using the some documentation on FreeBSD site.  I needed
> permission if I could post some of the documentation on my website.

The documentation is BSD licensed; see for example:
http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/LEGALNOTICE.html

So feel free to (re)use parts of it, under the terms of the BSD license.

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

Re: batch conversion of TeX

[Giorgos Keramidas]

On 2007-10-18 11:42, Murray Taylor <[EMAIL PROTECTED]> wrote:
>Giorgos Keramidas wrote:
>> I usually start by writing something like this in a Makefile:
>>
>> DOC = foo
>> SRC = $(DOC).tex
>> PDF = $(DOC).pdf
>>
>> PDFLATEX = pdflatex
>>
>> all: $(PDF)
>>
>> $(PDF): $(SRC)
>> $(PDFLATEX) $(SRC)
>> $(PDFLATEX) $(SRC)
>>
>> The two runs of $(PDFLATEX) are necessary to get cross-references
>> correct in documents with internal cross-references.
>
> the latex-mk port handles a lot of these functions
> /usr/ports/misc/latex-mk
>
> I uses it for all my docs
>
> make  # generates a DVI file and calls a viewer
> make ps
> make pdf
> make html # settable options re single page - multi page
> make draft-pdf# overprints DRAFT - use this if you are not using the
>
>   #  \usepackage{draftcopy} which gives you more
> flexibility
> make print# spools off to lp
> make clean
>
> And there are other available targets for the Make process.

Very interesting.  Thanks :)

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

Re: Why 7.0 is so late ?

[Giorgos Keramidas]

On 2007-10-17 18:20, Gueven Bay <[EMAIL PROTECTED]> wrote:
>2007/10/17, Wojciech Puchar <[EMAIL PROTECTED]>:
>>> while companies like Microsoft must publish in time to make the
>>> share holders
>>
>> not to mention that microsoft regularly has delays like year.
>>
>>> happy, FreeBSD can afford to wait until the programmers are
>>> convinced that their work is good enough for the public.
>
> Please,
> stop mentioning this company in every discussion that "seems" to
> "attack" *BSD in some way.

That's not a bad suggestion, in fact.  The development process of
Microsoft is not open to the world (like the one used by the FreeBSD
Project as a team), so there is no easy way to determine how similar
or different processes are behind the development and release process
of the two development teams.

Educated guesses can always be made, but let's try not to compare apples
to oranges too much :)

> Discussing M. - especially their development processes or business
> tactics - does not accomplish anything for us.

True, in a way.

> First acknowledge that FBSD 7 is late. The estimate was for 5 months
> ago. And if I say to someone I want to meet him at 8 o'clock but
> arrive 78 hours later, then I am late. Period.

The release of 7.0-RELEASE *is* late.  There are various reasons why
this has happened, but we are steadily getting there.  We now have a
RELENG_7 branch, and things are only "merged from current" after
explicit approval by the release-engineering team.  The branch is in the
hands of the RE team, and many issues which were plaguing "HEAD" during
the summer have been fixed now.

> So, but the original question was : What is not working _now_  at this
> moment so that 7 cannot be released _now_ ?
>
> The original question was not (I repeat NOT): Which  international
> conspiracy is holding up the developers of FreeBSD 7 from releasing?
> Without discussing M. or the Illuminati and U.F.O.s and with that
> guaranteeing that the original question will never be answered you can
> concentrate on answering OR you can take your "finger" from the reply
> button/menue point/whatever and wait that maybe a developer will read
> the original posters mail and answer it.
>
> With a curious eye waiting for a real answer to the original question...

Traditionally, "BSD" has released stuff "when it was ready" and not when
some marketting team decided that they wanted to release.  The FreeBSD
team has made genuine efforts towards changing this to a more timely
release schedule (18 months for a new "major" release), but there have
been some important bits of kernel and userland which were a bit
unstable and/or were in development until now.

Without treading on the feet of the release-engineering team, by writing
stuff which they have not approved, let me just say that we have made a
lot of positive progress towards a release since last June/July and we
expect getting a release during the last remaining months of 2007.

There's still a lot of work to do (i.e. ports to be compiled, tested,
fixed, or marked as "BROKEN" with the new gcc 4.X compiler suite), but
we're getting there.

- Giorgos

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