Re: Samba Shares, Unix Permissions, and Win2k clients
Am Dienstag, 18. Februar 2003 16:30 schrieb Doug MacFarlane: > Team: > > I need a pointer to a reference for how SAMBA handles permissions. > > I have a straight-forward install where the SAMBA server is the PDC, all > users have Unix accounts, and the SAMBA shares are setup to use the Unix > permissions. > > But in some cases, writing to a file sets permissions a 644 and in others, > 755. was it by using different shares on the server? Then you could have look in your /etc/samba/smb.conf. Inside the share sections you may (have) set a "create mask" : That is from smb.conf from one of my linux boxes running samba: [homes] comment = Home Directories writeable = Yes create mask = 0700 directory mask = 0700 [root] comment = root path = / writeable = Yes create mask = 0700 directory mask = 0700 ... (After changing you'll have to restart samba "/etc/init.d/samba restart".) You are also able to change settings by means of swat or webmin - if installed. Maybe you could also have a look in directory /usr/share/doc/samba(-doc, or -common) to get useful information nearby. Willi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: DNS + DHCP
Am Dienstag, 18. Februar 2003 15:30 schrieb Paul Johnson:
> On Mon, Feb 17, 2003 at 08:37:06AM -0600, Jeffrey L. Taylor wrote:
> > They work okay together using Dynamic DNS (not things like dyndns.org,
> > same name, different process). You can use TSIG (IIRC) to securely
> > authenticate updates.
>
> I tried before a couple times, and never could get it to work. I'd
> prefer to use IP ranges on my home network for authentication, but
> still didn't manage to make it work. Any pointers from anybody who
> has prodded it into production?
Bind 9 is able to receive dynamic updates by default. Important is
"allow-update" per zone. "localhost" is mandatory if dhcp is running on the
same machine. And : don't forget the reverse zone.
Some lines from a /etc/bind/named.conf.
...
zone "essignetz.de.local" {
type master;
file "/etc/bind/db.essignetz.de.local";
allow-update { 192.168.0.1; 127.0.0.1; };
};
zone "0.168.192.in-addr.arpa" {
type master;
file "/etc/bind/db.192.168.0";
allow-update { 192.168.0.1; 127.0.0.1; };
};
...
/etc/dhcp3/dhcp.conf
---
ddns-update-style interim;
option domain-name "essignetz.de.local";
default-lease-time 6;
max-lease-time 72;
# Interface eth1 :
subnet 192.168.0.0 netmask 255.255.255.0
{
authoritative;
}
---
If you need keying for security reasons, you should man "dhcpd.conf" and look
for the point "DYNAMIC DNS UPDATE SECURITY".
Willi
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Controlling swapping
Am Dienstag, 18. Februar 2003 19:53 schrieb Roy Pluschke: > On Tue, 18 Feb 2003 10:03:16 -0800 > Hi, on console you could try using "top". Watch the top-lines for overall statistics - including swapspace. "M" will sort by Memoryusage. "?" will show you a short help. "man top" (on commandline) will show more. "top b" makes it batch capable, so you may log top-data to a file for long time watching. Willi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Problem mit cyrus21-imapd
Am Donnerstag, 20. Februar 2003 17:25 schrieben Sie: > Wilfried Essig schrieb: > > Bist Du Dir da ganz sicher? Vielleicht hört ja der inetd > > auf Port 143 mit. > > Sieh mal in Deine /etc/inetd.conf, ob dort was für imap > > bzw. port 143 drin > > steht. > > Klar steht in inetd.conf ne Zeile für imap drin. Dort wird der > cyrus-imapd aufgerufen. Ist das etwa verkehrt? Kann leider erst heute > abend testen... > > Marko Hallo Marco, entweder kannst Du einen Dienst über inetd starten. Inetd ruft dann die angegebene Kommandozeile auf, wenn auf dem Port einer anständig fragt. Oder Du startest das Teil als Dämon über die Einträge in "/etc/rc[12345].d". Dann darfst Du aber keine entsprechende Zeile in der inetd.conf stehen haben. Sonst horcht nämlich inetd schon auf Port 143, und der cyrus bringt genau die von dir genannte Fehlermeldung. Also ich lasse ihn als Dämon-modus laufen. Da klappt es. Willi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: FreeS/WAN on PPPOE
Am Mittwoch, 26. Februar 2003 19:49 schrieb Curtis Vaughan: > Out DSL connection is through a PPPOE connect, unfortunately. Although > setting up multiple VPNs has generally been no problem, this time for > this office it has been a pain. The only reason we can think of now is > because this is first time we've dealt with a PPPOE connection. Has > anyone experienced similar problems? If so, were you able to overcome > it? Or, perhaps this isn't the problem at all. Hi, we connect several branch offices to our main office using freeswan. That makes no problems at all. But maybe your pppoe-Lines not comparable to ours in germany. We have pppoe on adsl lines (768 - 1500 Kbit/s down , 128 - 192 Kbits up) with dynamic IP-Addresses. Our main office has a 2Mbit Leased line with permanent ip-addresses. We running freeswan on woody and on potato. On woody we are using the woody-pakage (freeswan 1.96) , on potato we got freeswan directly from freeswan.org an compiled it ourselves (i think it's 1.92). Before going to ipsec : pppoe has less usable paketsizes. Are you able to get big pakets (without vpn) over your pppoe-line at all? Now ipsec : Maybe you should setup your pppoe - "ipsec.conf" with %defaultroute - if you've already having it like that. Have a look at some parts of our ipsec.conf : (left = branch office / rigth = main office) Part of ipsec.conf on a pppoe - Gateway : --- # /etc/ipsec.conf - FreeS/WAN IPsec configuration file # basic configuration config setup # THIS SETTING MUST BE CORRECT or almost nothing will work; # %defaultroute is okay for most simple cases. interfaces=%defaultroute # Debug-logging controls: "none" for (almost) none, "all" for lots. klipsdebug=none plutodebug=none # Use auto= parameters in conn descriptions to control startup. plutoload=%search plutostart=%search # Close down old connection when new one using same ID shows up. uniqueids=yes # defaults for subsequent connection descriptions conn %default keyingtries=0 disablearrivalcheck=no compress=yes authby=rsasig auto=start conn vh.n-lz.n leftsubnet=192.168.193.0/24 rightsubnet=193.158.106.0/24 also=vh.gw-lz.gw conn vh.gw-lz.n rightsubnet=193.158.106.0/24 also = vh.gw-lz.gw conn vh.n-lz.gw leftsubnet=192.168.193.0/24 also = vh.gw-lz.gw # "also" section : conn vh.gw-lz.gw left=%defaultroute [EMAIL PROTECTED] right=193.158.106.2 [EMAIL PROTECTED] rightnexthop=193.158.106.1 leftrsasigkey=0x0103bae5... rightrsasigkey=0x0103ad9... Part of ipsec.conf on central gateway : # basic configuration config setup interfaces="ipsec0=eth0" # Debug-logging controls: "none" for (almost) none, "all" for lots. klipsdebug=none plutodebug=none # Use auto= parameters in conn descriptions to control startup plutoload=%search plutostart=%search # Close down old connection when new one using same ID shows up. uniqueids=yes # defaults for subsequent connection descriptions conn %default # How persistent to be in (re)keying negotiations (0 means very). keyingtries=1 compress=yes right=193.158.106.2 [EMAIL PROTECTED] rightnexthop=193.158.106.1 authby=rsasig rightrsasigkey=0x0103a. auto=add # # Baum-vh-fw1 # conn vh.n-lz.n leftsubnet=192.168.193.0/24 rightsubnet=193.158.106.0/24 also=vh.gw-lz.gw conn vh.gw-lz.n rightsubnet=193.158.106.0/24 also=vh.gw-lz.gw conn vh.n-lz.gw leftsubnet=192.168.193.0/24 also=vh.gw-lz.gw # "also" section for vh : conn vh.gw-lz.gw left=%any [EMAIL PROTECTED] leftrsasigkey=0x0103... --- I hope this will help you. If not, you'll at least need to post your ipsec.conf-parts with the general section (cut your keys out) for the problematic connection. Maybe you could show the output of "ipsec barf" from both sites after trying to get a connection. (Thats about >>100K per site...) You could also have a look to freeswan.org. They have two sites with archives of their mailinglists. (Maybe that would be the best way, to get your freeswan-problem solved.) Willi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: adding eth1
Am Fre, 2003-02-28 um 21.06 schrieb Ina&Frank: > Debian Folks > > I have a next question. > > I wandered through the manpages and internet for a few hours but can't > seem to find a satisfactory answer for this one. > > I have an up and running IBM PC with Debian kernel 2.2.19 at the moment. > It has one NIC (rtl8139) which works fine. > I want to add a second identical card (eth1) but I am a bit confused how > to configure it. > Is modifying /etc/interfaces and /etc/ifstate sufficient? Where do I > tell the system what IRQ to use. And does that have to be another IRQ as > eth0? Or can they share it? > Or do I have to compile a new kernel? Or is there a config tool for it, > like gpmconfig for the mouse? > > I try to find out what to expect before trying it, so I have no > experiance yet. Hi Frank, if the second card is the same type, it will use the same driver. If it is a pci-Card you don't need to tell the driver module anything about the hardware. Try "ifconfig eth0" and "ifconfig eth1" to look for the Ports and interrupts. To get the IP to the hardware you will have to make allmost the same as for eth0 again for eth1 - maybe editing /etc/networks/interfaces if you are using ifup/ifdown. Willi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: AW: pppd problem...
Am Son, 2003-03-02 um 02.05 schrieb Christof Hurschler: > Sorry, but how do I get rid of the "tap0" default route? Dear Christof, maybe by deleting the line "gateway ..." from the interface-section of tap0 in your "/etc/network/interfaces"? IMHO pppd denies in many cases the setting of a default-route, if any other exists during startup of the ppp-connection. -- Wilfried Essig <[EMAIL PROTECTED]> Familie Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: crontab just wont stop mailing ... !
Am Sam, 2003-03-01 um 21.30 schrieb Dave Selby: > Anybody know where roots mail is kept ? Usually in file /var/mail/root. The other files in that directory are the mailstores from other users getting mail on your machine. (But not with cyrus imapd - and maybe other mailservers) Usually root shan't read his mail himself. Please look in file /etc/aliases for the name where the root-mail goes actually. -- Wilfried Essig <[EMAIL PROTECTED]> Familie Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Unidentified subject!
Am Son, 2003-03-02 um 03.54 schrieb debian_newbie: > Hello Everybody, > > How do I change the number of times I can boot up before fsck does a > complete file system check? It does it on my Woody machine every 20 > times. Also, I'm using ext3 filesystem. isn't it a journalized fs? I > thought journalized filesystems didn't have to be fscked? > tune2fs -c 0 -i 0 /dev/the-file-system-you-mounting > -- > debian_newbie, > Thanks -- Wilfried Essig <[EMAIL PROTECTED]> -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Network config help needed
Am Son, 2003-03-02 um 13.15 schrieb [EMAIL PROTECTED]: > Please help me > I'm trying to connect two pc's, but I allways get the message > "unable to connect to remote host: No route to host". What am I > missing? > > My config for pc1 looks like this: > ifconfig eth0 192.168.0.1 netmask 255.255.255.0 up > rout add -host 192.168.0.2 eth0 You don't need this special route to host ".2" - it's alraedy fullfilled with the standard-route to the network. But it should not disturb. > "route -n" output: > 192.168.0.2 0.0.0.0 255.255.255.255 UH 0 0 0 eth0 > 192.168.0.0 0.0.0.0 255.255.255.0UH 0 0 0 eth0 > > /etc/hosts > domainhome > search home > nameserver 10.0.0.1 > nameserver 10.0.0.2 Where is the route to your nameservers? -- Wilfried Essig <[EMAIL PROTECTED]> Familie Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: getting the initrd to load
Am Don, 2003-03-06 um 01.33 schrieb Ian Melnick: > > You can use romfs or even ext2/ext3. Check out the MKIMAGE setting > > in mkinitrd.conf. > > Okay, I used genromfs (i think) according to the example in your man page. > Lilo doesn't load it, and on startup, I still get a message: > "RAMDISK: Couldn't find valid ram disk image starting at 0." > It's still in /boot; in lilo.conf it says: initrd=/boot/initrd-2.4.20 > > What now? > Thanks Hi, looks like your initrd is no valid disk (more likely) or your kernel is compiled without support for "cramfs" in filesystems / blockdevices (less likely i think). If initrd is not in place the message should look something like this : > request-module[block-major-3]: Root fs not mounted > VFS: Cannot open root device "301" or 03:01 > Please append a correct "root=" boot option > Kernel panic: VFS: Unable to mount root FS on 03:01 Try "mkinitrd". I used it many times - and it runs. Use it after installing the new kernel and the new modules. If your current kernel ist not the same number than your new one, you will have to extend the mkintrd-command with the path of your new modules-dir (usually /lib/modules/2.4.20 for a kernel 2.4.20) : mkinitrd -o your-initrd-name /lib/modules/2.4.20 -- Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: How to gather information about a open port?
Am Don, 2003-03-06 um 10.45 schrieb Qian Gong: > Hi, > > I use nmap to detect the security of my friend's win2k machine and find > a open tcp port 1039. We tried several days but can not find any > information about this port. Is it possible to use some Linux tools to > probe this port and gather information during communication with this > port? Thanks a lot for any suggestions. > > Qian There is a freeware tool to run on the windows machine - telling you in addition to the open ports the programs behind. It's called "Active Ports". You can get on : http://www.ntutility.com/freeware.html -- Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Unidentified subject!
Am Don, 2003-03-06 um 14.41 schrieb Jerry Van Brimmer: > tune2fs -c 0 -i 0 /dev/the-file-system-you-mounting > > > When I run #tune2fs -i 30d /dev/hda > > I get: > > tune2fs 1.27 (8-Mar-2002) > tune2fs: Bad magic number in super-block while trying to open /dev/hda > Couldn't find valid filesystem superblock. > debian:/home/jerry# > Try it with "the-file-system-you-mounting" - not with "the_disk_on_witch_ the_"the-file-system-you-mounting"_is" -- Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: where does "root" mail go ?
Am Don, 2003-03-06 um 19.20 schrieb Dave Selby: > When an application sends mail to ''root" where does it go ? because I cant > find it ! All mail I have ever accessed is via the web with kmail. > > Is it supposed to be posted to ... > /var/mail/ ? Yes. That's the directory for local delivery. In many cases the mail to root will end up there. Please look in /etc/aliases to see which user is getting the root mail (root is not expected to read his mail himself - because of security considerations) Next you should have look in directory /home/target-user-for-the-root-mail. Is there a file named ".forward"? Look for the destination adress in .forward. If there is nothing to find, you need to post with MTA (Sendmail,...) and with local-delivery-agent (procmail, ...) you are using. Do you use cyrus imapd? -- Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Reason for system reboot?
Am Mon, 2003-03-10 um 10.01 schrieb Jan Johansson: > Is there someway to find out what caused a system reboot? I have the > following in my "last" > > argus:~# last -20 > root pts/0192.168.221.207 Mon Mar 10 08:11 still logged > in > reboot system boot 2.4.18-bf2.4 Sun Mar 9 17:49 (16:06) > > root pts/0192.168.221.207 Fri Mar 7 16:31 - crash > (2+01:18) > But i cant find anything in the logs which shows why the system was > rebooted. So i'am. Looks something garbaged. Is that really wath you are finding in your logs? Try have a look to /var/log/syslog with an editor. > What worries me more is that I should be the only one with > root access to this system, which Is a minimalistic Debian/Testing doing > not much else then running shorewall. Do you have pressing ctrl-alt-del for restarting system active? Because i have some win-nt/2000/xp-boxes with Netware and linux nearby another i usually disable this. (Makes much less unnessecary restarts on the linux machines.) -- Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Kernel panic: No init found.
Am Mit, 2003-03-12 um 19.57 schrieb Reid Mumford: > On Wed, 12 Mar 2003, sean finney wrote: > > > On Wed, Mar 12, 2003 at 01:09:50PM -0500, Reid Mumford wrote: > > > EXT2-fs warning: checktime reached, running e2fsck is recommended > > > VFS: Mounted root (ext2 filesystem). > > > VFS: Mounted root (ext2 filesystem) readonly. > > > change_root: old root has d_count=1 > > > Trying to unmount old root ... okay > > > Freeing unused kernel memory: 156k freed > > > Kernel panic: No init found. Try passing init= option to kernel > > > > > > Any ideas on how to save this system without having to reinstall? > > After doing a disk check you could do this : Do you have a healthy /etc/inittab? Does your kernel use initrd.img? Is it healthy? Maybe you could boot your system with a rescue-system (using nearby kernel version) making "chroot /mnt/yourbootdrive /bin/bash". There you could try rerun lilo. Next you could look inside chroot how much of your system is damaged. Maybe you would be able to use commandline dpkg to overwrite/reinstall defective packages. But that looks like much work. Consider backing up your data and reinstalling. It may be less effort. Willi -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: AARRGGHH: Samba + WinXP Home
Dear David,
David Purton sagte:
> We Just bought a new server at work and its running linux + samba.
> Most seems to run fine, except I can't make the WinXP boxes remember
> their passwords for reconnecting to shares at login.
I assume you are able to connect to samba "by hand" (by explorer and
mouse). Maybe i could ask you if you know that "smbpasswd" on linux
console is necessary to change samba-passwords?
Are you asked for username/password when making the connect by hand? Is
your samba login and password the same as your locals at winxp? If not, it
would be much better make them the same.
If the problem is only the automatic login at startup i would expect, that
your Workstation won't find the server at once. In that case you should
make your samba a wins server. Have also a look at your dns. Maybe you
should create file "lmhosts" out of "lmhosts.sam" (editing name and adress
of your server into it) and set "use lmhosts" inside the properties of yor
tcp/ip-stack.
Do you get your ip-address by dhcp? then you could give your wxx helpful
tips by your dhcp.conf :
---
subnet 192.168.0.0 netmask 255.255.255.0
{
authoritative;
range 192.168.0.20 192.168.0.99;
option domain-name-servers 192.168.0.1;
option routers 192.168.0.1;
# this to lines :
option netbios-name-servers 192.168.0.1;
option netbios-node-type 8;
# !!
option broadcast-address 192.168.0.255;
}
---
Maybe, my smb.conf is more worthy for you. It runs with winxp-home and w95
very clean.
# Global parameters
[global]
interfaces = 192.168.0.0/255.255.255.0
passwd program = /usr/bin/passwd %u
debug level = 1
passwd chat = *Enter\snew\sUNIX\spassword:* %n\n
*Retype\snew\sUNIX\spassword:* %n\n .
read prediction = yes
dns proxy = No
encrypt passwords = Yes
server string = %h server (Samba %v)
workgroup = Essignetz
syslog = 0
getwd cache = yes
socket options = TCP_NODELAY IPTOS_LOWDELAY
bind interfaces only = Yes
wins support = true
hosts allow = 192.168.0.0/255.255.0.0
[homes]
comment = Home Directories
writeable = Yes
create mask = 0700
directory mask = 0700
[root]
comment = root
path = /
writeable = Yes
create mask = 0700
directory mask = 0700
[printers]
comment = All Printers
path = /tmp
create mask = 0700
printable = Yes
browseable = No
> If you, O LORD, kept a record of sins, O Lord, who could stand?
> Psalm 130:3
Do you believe in Christ? Then you will know the answer ;-)
--
MfG / Yours
Wilfried Essig
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: network problem: configuration/DNS? cannot access internal machine using our external IP
Dear Jerome, looks like you have two problems : First : Ping works - tracroute not Are using your Wxx and Linx-Boxes same DNS? If yes, try "traceroute -n EXT-IP" on your Linux boxes. (Ping doesn't a try to get the name for the given ip-address, but traceroute does - and DNS-timeouts are rather long). You tried to disable firewall. That didn't help, because it seems not to be the problem with the different behavior of ping and traceroute on your linux boxes. You should have all internal machines get any internetnames resolved into ip-addresses, including xx.dyndns.org! Now, if all your Workstations and your server should be able to make dns well, we'll come to next problem : Looks like your router does not do the portforwarding for pakets coming from the internal net. Maybe its a router problem, maybe firewall. Try switching firewall off again. Maybe your router has a "switch" to make him forward pakets from the inside with destination adress "EXT-IP" back to the inside - with destination is ip of the server. To latest notes : i don't think the xx.dyndns.org entry in hosts does do that much for the machines on your internal network. But try it without. Linux itself does not need to know the name of EXT-IP to run smooth. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Problems with IMAP
On Tue, 2003-09-30 at 06:02, Roberto Sanchez wrote: > What is giving me the problem is this: > > smtpd_recipient_restrictions = permit_mynetworks, check_sender_access > hash:/etc/postfix/access, permit_tls_clientcerts, > reject_unauth_destination, reject All postfix-smtpd are using this line(s) in main.cf. I think, you'll need different configs for smtp on port 25 and the other smtpd listening on port 465. There are at least two ways to get it. You could config a second (or even more) instances of postfix. The other way is to config the different smtpd in master.cf : Change your smtpd-entry for port 465 in master.cf like this : -- smtps inet n - n - - smtpd -o smtpd_tls_wrappermode=yes -o smtpd_sasl_auth_enable=yes -o smtpd_client_restrictions= -o smtpd_helo_restrictions= -o smtpd_sender_restrictions= -o smtpd_recipient_restrictions=permit_mynetworks,reject -o smtpd_recipient_restrictions = permit_mynetworks, check_sender_access hash:/etc/postfix/access, permit_tls_clientcerts, reject_unauth_destination, reject -o smtpd_what_you_need... -- Please note : 1. the "empty" lines like "-o local_recipient_maps=" are for switching off according, but for this task unwanted, lines in main.cf. 2. Don't use the above directly : it's only an example, to show you syntax and possibilities. I'm using the postfix 2.0 backports from http://people.debian.org/~hmh . Maybe not all will work the same on postfix 1.x. Have a look to your postfix doc (Filter_Readme?) Hope this will show you a direction... Yours, Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: postfix pop-before-smtp problem
On Tue, 2003-09-30 at 18:02, Hans Steinraht wrote: > smtpd_recipient_restrictions = > permit_mynetworks,reject_non_fqdn_recipient,check_client_access > hash:/var/lib/pop-before-smtp/hosts,check_relay_domains > > After restarting Postfix all looks fine but by receiving the first email > Postfix starts to complaing with: > > Sep 30 17:52:12 esteta postfix/smtpd[6216]: fatal: open database > /var/lib/pop-before-smtp/hosts.db: Invalid argument - Has your postfix enough rights to access your hosts.db? - Is your smtpd running chrooted? Is your hosts.db available in the chroot-jail? - Is hosts.db really db-format? Yours Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: Spamd error
Am Son, 2003-07-27 um 11.09 schrieb J. Zidar: > When I try to daemonize spamd I get the folowing error: > Cannot assign requested address at /usr/sbin/spamd line 156. > > The relative line from the spamd script is: > bind(Server, sockaddr_in($port, $addr)) || die "bind: $!"; > Since spamd is not running I assume that even spamc won't run. I'm using > fetchmail/procmail to fetch and process my mail. One of the lines in my > .procmailrc explicitly run spamc, but it is not working because spam has > (once again) poluted my inbox. > > Or should I just use plain spamassassin in my procmail recipe? > > Any push in the right direction would be great. Did you CHECK "spamd" is not already running? ( using "ps ax") A running spamd should listen on port 783. Take a look with "netstat -pln" > > Jernej Zidar -- Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: SQL Help
On Thu, 2004-04-08 at 11:15, Saqib Qureshi wrote: > but actually i want to fatch each year record in a separate variable and in > a single query how it is possible. Hi, try this : select a2.zon, b2.astyer, count(a2.ntn), b3.astyer, count(a3.ntn) from ztrmtrext a2, rtnmtr b2, ztrmtrext a3, rtnmtr b3 where a2.ntn = b2.ntn and b2.astdat is null and b2.astyer='2002' and a3.ntn = b3.ntn and b3.astdat is null and b3.astyer='2003' and a2.zon = a3.zon group by a2.zon, b2.astyer, b3.astyer ; Please note, that you will only get lines listet that have the "zon"-Values in any year asked for. You will have to use "outer joins" if not all "zon"-Values are in any year. For details on what an "outer join" is, you can look at http://www.1keydata.com/sql/sqlouterjoin.html or http://www.adp-gmbh.ch/ora/sql/outer_join.html . Wilfried Essig -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
