This site is now 100% read-only, and retired.

XML Logo

Posted by muondude on Wed 6 Jul 2011 at 19:49
Tags: ,
I ran aptitude and tried to install the latest Bind9 security update.
(1:9.7.3.dfsg-1~squeeze3)

The update failed and left my bind install in "iF" state - and I can't seem to repair it. The config files even seem to have gone away from /etc!!
I posted a bug report: <http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=632815>;
Bind9 has always been funky on this PPC Mac Mini.

Any suggestions about how to recover. I'd even settle for removing bind9 and reinstalling the previous version. I am pretty sure I have a backup of the config files since I backup /etc 2x/month.
2011.07.10
I tried to fix this and got:
>apt-get -f install
Reading package lists... Done
Building dependency tree       
Reading state information... Done
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.
1 not fully installed or removed.
After this operation, 0 B of additional disk space will be used.
Setting up bind9 (1:9.7.3.dfsg-1~squeeze3) ...
Stopping domain name service...: bind9Awakening mail retriever agent: fetchmail failed!
run-parts: /etc/resolvconf/update-libc.d/fetchmail exited with return code 1
run-parts: /etc/resolvconf/update.d/libc exited with return code 1
invoke-rc.d: initscript bind9, action "restart" failed.
dpkg: error processing bind9 (--configure):
 subprocess installed post-installation script returned error exit status 1
configured to not write apport reports
                                      Errors were encountered while processing:
 bind9
E: Sub-process /usr/bin/dpkg returned an error code (1)

2011.07.10.13.23.45
I purged fetchmail and bind9 seemed to get fixed... ??
>apt-get purge fetchmail
Reading package lists... Done
Building dependency tree       
Reading state information... Done
The following packages were automatically installed and are no longer required:
  blt python-tk
Use 'apt-get autoremove' to remove them.
The following packages will be REMOVED:
  fetchmail* fetchmailconf*
0 upgraded, 0 newly installed, 2 to remove and 0 not upgraded.
1 not fully installed or removed.
After this operation, 2,867 kB disk space will be freed.
Do you want to continue [Y/n]? 
(Reading database ... 51812 files and directories currently installed.)
Removing fetchmailconf ...
Purging configuration files for fetchmailconf ...
Removing fetchmail ...
Pidfile not found! Is fetchmail running? ... failed!
Purging configuration files for fetchmail ...
Processing triggers for man-db ...
Setting up bind9 (1:9.7.3.dfsg-1~squeeze3) ...
Stopping domain name service...: bind9 waiting for pid 8891 to die.
Starting domain name service...: bind9
...

Bind still seems to hang at restart but now it shows:
>dpkg -l bind9
Desired=Unknown/Install/Remove/Purge/Hold
| Status=Not/Inst/Conf-files/Unpacked/halF-conf/Half-inst/trig-aWait/Trig-pend
|/ Err?=(none)/Reinst-required (Status,Err: uppercase=bad)
||/ Name                        Version                     Description
+++-===========================-===========================-======================================================================
ii  bind9                       1:9.7.3.dfsg-1~squeeze3     Internet Domain Name Server


TIA

-- Sam

 

Posted by muondude on Sun 26 Apr 2009 at 19:44
Tags: none.
I have been using Apcupsd with the APC UPS: Smart-UPS Model Number: DLA1500RM2U. The monitoring was done from a old Mac G4 which I am retiring and I'd like to move the monitoring to a Mac Mini (PPC) running lenny.

Both Apcupsd and NUT are available on lenny.

I have one UPS that I monitor and my monitoring/control needs are modest.
My DSL model, FireWall, and Web/Email and Backup servers are all that are connected (yes I know this UPS is probably overkill).

Any suggestions about which is the easiest to setup and maintain for a simple situation? Other pro/con issues?

TIA.

-- Sam

 

Posted by muondude on Thu 23 Apr 2009 at 20:10
I recently upgraded from etch to lenny. I then found that bind "seems" to be working but is acting strange. Here's some output from the process start and stop and the /etc/bind9 config file list. Note that the "start" after stopping can take up to 5 minutes.

I'm running on a Mac Mini (PPC version). All other processes (mail, web) seem to be working okay.

Thoughts on why this is occurring?

-- Sam



ls -l /etc/bind
-rw-r--r-- 1 root root 237 2007-01-29 06:00 db.0
-rw-r--r-- 1 root root 271 2007-01-29 06:00 db.127
-rw-r--r-- 1 root bind 959 2009-03-05 18:24 db.168.192
-rw-r--r-- 1 root root 237 2007-01-29 06:00 db.255
-rw-r--r-- 1 root root 353 2007-01-29 06:00 db.empty
-rw-r--r-- 1 root bind 1337 2009-04-22 09:39 db.gasster
-rw-r--r-- 1 root root 270 2009-03-18 20:39 db.local
-rw-r--r-- 1 root root 2878 2009-03-18 20:39 db.root
-rw-r--r-- 1 root bind 907 2009-03-18 20:39 named.conf
-rw-r--r-- 1 root bind 851 2008-01-03 15:08 named.conf.local
-rw-r--r-- 1 root bind 1253 2009-04-22 08:47 named.conf.options
-rw-r--r-- 1 root bind 572 2009-03-18 20:39 named.conf.options.dpkg-dist
-rw-r----- 1 bind bind 77 2007-04-16 22:44 rndc.key
-rw-r--r-- 1 root root 1317 2007-01-29 06:00 zones.rfc1918


This entry has been truncated read the full entry.

 

Posted by muondude on Mon 24 Dec 2007 at 18:19
Tags: none.
Over the last couple of days I've been experiencing a DOS attack on my mail server. I'm running Debian 3, exim3. The mainlog file is filled with the following:
...
2007-12-24 10:06:01 verify failed for SMTP recipient aaborkar@filemakerauction.com from <> H=web6.ci06.de (ci06.de) [193.143.122.10]
2007-12-24 10:06:02 verify failed for SMTP recipient harald@filemakerauction.com from <> H=(scmgateway1.reviewjournal.com) [12.9.217.24]
2007-12-24 10:06:04 Connection from 62.190.15.43 refused: too many connections
2007-12-24 10:06:04 Connection from 62.190.15.43 refused: too many connections
2007-12-24 10:06:05 verify failed for SMTP recipient eifmanl@filemakerauction.com from <> H=(mail2.e-servicesgroup.com) [65.183.1.213]
2007-12-24 10:06:05 Connection from 67.90.241.200 refused: too many connections
2007-12-24 10:06:05 Connection from 70.84.16.194 refused: too many connections
...
And netstat gives:
...
Active Internet connections (w/o servers)
Proto Recv-Q Send-Q Local Address           Foreign Address         State      
tcp        0      1 192.168.2.2:3050        148.245.120.60:113      SYN_SENT   
tcp        0      0 192.168.2.2:25          148.245.120.60:31146    ESTABLISHED
tcp        0      1 192.168.2.2:3049        148.245.120.60:113      SYN_SENT   
tcp        0      0 192.168.2.2:25          148.245.120.60:31144    ESTABLISHED
tcp        0      1 192.168.2.2:3047        209.82.14.94:113        SYN_SENT   
tcp        0      0 192.168.2.2:25          209.82.14.94:30136      ESTABLISHED
tcp        0      0 192.168.2.2:25          213.201.175.98:1574     TIME_WAIT  
tcp        0      1 192.168.2.2:3045        217.154.210.139:113     SYN_SENT   
tcp        0      0 192.168.2.2:25          217.154.210.139:43850   ESTABLISHED
tcp        0      1 192.168.2.2:3044        217.154.210.139:113     SYN_SENT   
tcp        0      0 192.168.2.2:25          217.154.210.139:43231   ESTABLISHED
tcp        0      1 192.168.2.2:3043        217.154.210.139:113     SYN_SENT   
tcp        0      0 192.168.2.2:25          217.154.210.139:43160   ESTABLISHED
tcp        0      1 192.168.2.2:3042        80.219.58.34:113        SYN_SENT   
tcp        0      0 192.168.2.2:25          80.219.58.34:1605       ESTABLISHED
tcp        0      1 192.168.2.2:3039        62.80.22.166:113        SYN_SENT   
tcp        0      0 192.168.2.2:25          62.80.22.166:2006       ESTABLISHED
... snip...
48 total connections on tcp & smtp. Ugh!
I've tried googling for help, but nothing I could 'grok' or figure out. Also, it seems that exim4 might have some features to handle this, but I'm unsure about this and frankly afraid to break the server with the update doing on my own (yes I know it is hosed so how much more could I screw it up ;-) ). Any suggestions would be greatly appreciated. If there is a kind person willing to help I can contact you on a side channel (phone or IM). Thanks.

 

Posted by muondude on Wed 25 Apr 2007 at 16:28
Tags: none.
I need to run apt/aptitude behind a firewall that requires a username
and password.

All the examples I've seen require you to embed the username and password
in the proxy specification (either in apt.conf or an environment variable
not a good idea).

For example:

export http_proxy=http://username:password@proxyserver.net:port/

export ftp_proxy=http://username:password@proxyserver.net:port/

or you can configure this in the apt.conf file using a similar format
but it still requires one to include the u/p.

Is there a way to have apt/aptitude prompt the user for the username/password?

 

Posted by muondude on Tue 10 Apr 2007 at 18:51
Tags: none.
Debian Stable (etch):

I thought there was a meta-package (virtual package) for installing all the C/C++ development libraries and header files, but can't seem to find it.

What is the currently recommended way to get all those files (libraries and headers) installed.

I recently had a problem under etch where we couldn't build an app and none of the headers/libs were installed and it seemed like a random process to figure out what we needed.

 

Posted by muondude on Sat 16 Sep 2006 at 22:00
Tags: none.
I was looking at my procmail log file and noticed some spamassassin error messages. I did a little google search and I may have a broken install.

I am running Debian 3.1 which was an upgrade from Debian 2.

I've include some other information below.

the last part of my .procmail file looks like:


# removed -P at EOL
:0fw
| spamassassin

:0:
* X-Spam-Status: yes
spam

# deliver to local maildir mailbox
:0:
/home/sgasster/Maildir/


***
I'm wondering about how to fix this.
Is it simply a matter of telling spamassassin to look in the correct
config directory? I wasn't sure how to do this or where it should go?
In the spamassassin command in the .procmail file should I point to
the /usr/share/spamassassin directory, since it may be defaulting to
the "old - incorrect" /etc/spamassassin directory?

***


&gt; spamassassin -VD
SpamAssassin version 3.0.3
running on Perl version 5.8.4

In /etc I have:

ll /etc/mail
total 0
lrwxrwxrwx 1 root root 15 Jun 14 22:08 spamassassin -&gt; ../spamassassin

ls /etc/spamassassin/
10_misc.cf 25_body_tests_es.cf 30_text_pl.cf 65_debian.cf local.cf.dpkg-old
20_body_tests.cf 25_body_tests_pl.cf 40_spam_phrases.cf init.pre user_prefs.template
20_head_tests.cf 25_head_tests_pl.cf 50_scores.cf local.cf
20_uri_tests.cf 30_text_es.cf 60_whitelist.cf local.cf.bck

&gt; whereis spamassassin
spamassassin: /usr/bin/spamassassin /etc/spamassassin /usr/share/spamassassin /usr/share/man/man1/spamassassin.1p.gz




&gt; ls /usr/share/spamassassin/
10_misc.cf 20_fake_helo_tests.cf 20_ratware.cf 25_uribl.cf 60_whitelist.cf
20_anti_ratware.cf 20_head_tests.cf 20_uri_tests.cf 30_text_de.cf 65_debian.cf
20_body_tests.cf 20_html_tests.cf 23_bayes.cf 30_text_fr.cf languages
20_compensate.cf 20_meta_tests.cf 25_body_tests_es.cf 30_text_nl.cf triplets.txt
20_dnsbl_tests.cf 20_phrases.cf 25_hashcash.cf 30_text_pl.cf user_prefs.template
20_drugs.cf 20_porn.cf 25_spf.cf 50_scores.cf


*** so it seems like I have two sets of config files!

A sample of the procmail logs looks like:

...snip...
Failed to run LINES_OF_YELLING_2 SpamAssassin test, skipping:
(Can't locate object method "check_for_num_yelling_lines" via package "Mail::SpamAssassin::PerMsgStatus" at /usr/share/perl5/Mail/SpamAssassin/PerMsgStatus.pm line 2340.
)
Failed to run BASE64_ENC_TEXT SpamAssassin test, skipping:
(Can't locate object method "check_for_base64_enc_text" via package "Mail::SpamAssassin::PerMsgStatus" at /usr/share/perl5/Mail/SpamAssassin/PerMsgStatus.pm line 2340.
)
Failed to run RAZOR_CHECK SpamAssassin test, skipping:
(Can't locate object method "check_razor" via package "Mail::SpamAssassin::PerMsgStatus" at /usr/share/perl5/Mail/SpamAssassin/PerMsgStatus.pm line 2340.
)
procmail: [28536] Sat Sep 16 13:27:43 2006
procmail: Match on "X-Spam-Status: yes"
procmail: Locking "spam.lock"
procmail: Assigning "LASTFOLDER=spam"
procmail: Opening "spam"
procmail: Acquiring kernel-lock
procmail: Unlocking "spam.lock"
...snip...

TAI...

 

Posted by muondude on Fri 12 May 2006 at 18:56
Tags: none.
I've been looking at the Dell PowerEdge 850 rack servers.
I'm interest in RAID and they use the CERC SATA RAID controller.

Anyone know if there are drivers under Debian for this device?
I've done some searching but have not found any clear documentation that helps me understand this controller and if it works under Debian.

 

Posted by muondude on Thu 22 Sep 2005 at 04:17
Tags: none.
Some handy command line scripts that I have often found useful.
Search for files with extension .tex starting in current directory and
below, print result to stdout, which then get passed to script; 
./bin/test.sh (just a dummy script I created to print the filename out):

  find . -path '*.tex' -print -exec ./bin/test.sh {} \;


Command Line file renaming: rename a group of files with extension .html to .shtml extensions:

 perl -e 'for (@ARGV) { ($new=$_) =~ s/(.+)(.)\.html$/$1$2.shtml/; rename $_, $new unless -e $new }' *.html

Convert a bunch of file names from UPPER case to lower case

 perl -e 'for (@ARGV) { ($new=$_) =~ tr/[A-Z]/[a-z]/; rename $_, $new unless -e $new }' *.html

Command line file editing:

  perl -p -i -e 's/\.html/\.shtml/g' j*.shtml

  perl -p -i -e 's/VLINK="\#FF0066"/VLINK="\#009900"/g' *.shtml

list lines with specified text (kinda like grep):

  perl -n -e 'print if /VLINK="\#FF0066"/;' *.shtml

Edit the file in place and make a backkup copy with extension .orig
this one line command replaces the Mac newline with a Unix newline:

  perl -p -i.orig -e 's/\r/\n/g' testfile.txt

 

Posted by muondude on Tue 20 Sep 2005 at 06:26
Tags: none.
I upgraded from Woody to Sarge and I'm trying to find a 'safe' way to remove all the X-windows related packages/files and keep the server running.

I only access the server from the console or via SSH and so don't need X11.