Hi guys, any CUPS / Samba experts out there?
The CUPS printing guide is a horrendous maze of options. I've got nearly there but I've got one thing wrong and I don't know where to look.
I set up our two printers (HP K5400s) using the GUI printer tool. Both print the test page just fine. I set up the [printers] and [print$] shares and can browse to the printers from Windows (XP for me) I managed to get the postscript drivers installed with cupsaddsmb and it installs on the XP client just fine.
..... sadly it just prints the postscript instead of rendering it. I guess the printer queue is in raw mode but I can't see where to change it. I vaguely remember foomatic and such from ages back. Can someone point me at the right part of the setup to fix this please?
Kevin Thorpe wrote:
Hi guys, any CUPS / Samba experts out there?
The CUPS printing guide is a horrendous maze of options. I've got nearly there but I've got one thing wrong and I don't know where to look.
I set up our two printers (HP K5400s) using the GUI printer tool. Both
The GUI printer tool - is that from the o/s, or, from the server, did you browse to http://localhost:631/? From the latter, you can add printer->choose driver. <snip> mark
On Wed, Jun 22, 2011 at 3:44 PM, m.roth@5-cent.us wrote:
Kevin Thorpe wrote:
Hi guys, any CUPS / Samba experts out there?
The CUPS printing guide is a horrendous maze of options. I've got nearly there but I've got one thing wrong and I don't know where to look.
I set up our two printers (HP K5400s) using the GUI printer tool. Both
The GUI printer tool - is that from the o/s, or, from the server, did you browse to http://localhost:631/? From the latter, you can add printer->choose driver.
Both the KDE printer manager and the CUPS manager on :631 create the same printer setup.
Dear Kevin,
On Wed, 22 Jun 2011 15:16:54 +0100 Kevin Thorpe kevin.thorpe@pibenchmark.com wrote:
Hi guys, any CUPS / Samba experts out there?
The CUPS printing guide is a horrendous maze of options. I've got nearly there but I've got one thing wrong and I don't know where to look.
you have to do the following:
In CUPS https://localhost:631/ setup your printers and set the printer driver for each printer to local raw printer
In /etc/samba/smb.conf make sure you have the following settings
[main] printing = cups
# printer share for your windows drivers [print$] comment = printer drivers path = /var/lib/samba/somewhere read only = yes guest ok = no write list = @somegroup, john
mkdir /var/lib/samba/somewhere/{COLOR,IA64,W32ALPHA,W32MIPS,W32PPC,W32X86,WIN40,x64}
# set permissions chown -R john:somegroup /var/lib/samba/somewhere chmod 775 /var/lib/samba/somewhere
or if you use ACL setfacl --recursive -m g:somegroup:rwx /var/lib/samba/somewhere etc.
Grant the SePrintOperatorPrivilege privilege
net rpc rights grant "<domain|hostname<user|group>" \ SePrintOperatorPrivilege -U <hostname>/root
In windows open the print management tool and install the device drivers there.
HTH, Brgds
On Wed, Jun 22, 2011 at 4:02 PM, Benjamin Hackl b.hackl@focusmr.com wrote:
Dear Kevin,
On Wed, 22 Jun 2011 15:16:54 +0100 Kevin Thorpe kevin.thorpe@pibenchmark.com wrote:
Hi guys, any CUPS / Samba experts out there?
you have to do the following:
In CUPS https://localhost:631/ setup your printers and set the printer driver for each printer to local raw printer
Actually that's the opposite of what I want. I want to use the generic postscript driver in Windows and get the server to RIP for me.
Kevin Thorpe wrote:
On Wed, Jun 22, 2011 at 4:02 PM, Benjamin Hackl b.hackl@focusmr.com wrote:
Dear Kevin,
On Wed, 22 Jun 2011 15:16:54 +0100 Kevin Thorpe kevin.thorpe@pibenchmark.com wrote:
Hi guys, any CUPS / Samba experts out there?
you have to do the following:
In CUPS https://localhost:631/ setup your printers and set the printer driver for each printer to local raw printer
Actually that's the opposite of what I want. I want to use the generic postscript driver in Windows and get the server to RIP for me.
And you say that, using localhost:631, you did add printer, then selected make and model, and chose one of the drivers, and it still doesn't do the right thing?
mark
On Wed, 22 Jun 2011 15:16:54 +0100 Kevin Thorpe kevin.thorpe@pibenchmark.com wrote:
Hi guys, any CUPS / Samba experts out there?
In CUPS https://localhost:631/ setup your printers and set the printer driver for each printer to local raw printer
And you say that, using localhost:631, you did add printer, then selected make and model, and chose one of the drivers, and it still doesn't do the right thing?
That sets the printer up just fine in Linux. I can print to that fine with Open Office. The problem I have is that a generic PS driver in Windows to the SMB printer outputs the raw PS script instead of rendering which is odd because the CUPS test page is a postscript file.
Kevin Thorpe wrote:
On Wed, 22 Jun 2011 15:16:54 +0100 Kevin Thorpe kevin.thorpe@pibenchmark.com wrote:
Hi guys, any CUPS / Samba experts out there?
In CUPS https://localhost:631/ setup your printers and set the printer driver for each printer to local raw printer
And you say that, using localhost:631, you did add printer, then selected make and model, and chose one of the drivers, and it still doesn't do the right thing?
That sets the printer up just fine in Linux. I can print to that fine with Open Office. The problem I have is that a generic PS driver in Windows to the SMB printer outputs the raw PS script instead of rendering which is odd because the CUPS test page is a postscript file.
Hmmm... maybe we're looking at this the wrong way: what is the setup on the *Windows* side - what driver does that use?
mark
In CUPS https://localhost:631/ setup your printers and set the printer driver for each printer to local raw printer
And you say that, using localhost:631, you did add printer, then selected make and model, and chose one of the drivers, and it still doesn't do the right thing?
That sets the printer up just fine in Linux. I can print to that fine with Open Office. The problem I have is that a generic PS driver in Windows to the SMB printer outputs the raw PS script instead of rendering which is odd because the CUPS test page is a postscript file.
Hmmm... maybe we're looking at this the wrong way: what is the setup on the *Windows* side - what driver does that use?
I put together the cups PS driver (cupsps6.dll et.al.) and the postscript driver from HP laserjet PS (pscript5.dll et.al.) as described in the CUPS HOWTO then used cupsaddsmb to make it available to the workstations. This driver installs automatically just fine when I connect to the printer share and quite correctly produces a postscript file - unfortunately samba+CUPS don't recognise the need to render it. I haven't got cups options = raw in my smb.conf so it shouldn't bypass the CUPS filtering (I don't think)
Hi guys, any CUPS / Samba experts out there?
Bah! I saw that lpr could print PS but smbclient coudn't so in desperation I restarted samba. Now it prints PS. Why is completely beyond me. That's why I couldn't work out what I'd done wrong - I hadn't.
Kevin Thorpe wrote:
> Hi guys, any CUPS / Samba experts out there?
Bah! I saw that lpr could print PS but smbclient coudn't so in desperation I restarted samba. Now it prints PS. Why is completely beyond me. That's why I couldn't work out what I'd done wrong - I hadn't.
Did you fix cups, and not restart samba? If so, I'll be the info was cached, and had to be reread to know what to do.
mark
I am getting what I believe to be inconsistent replies/results from my CentOs machine, trying to ping URLs. Pinging by URL fails the DNS lookup, while pinging the IP address (that should have been returned) works. The DNS lookup also fails NSLOOKUP, but interestingly, dig and wget return A records with the correct IP.
Robert Henrichs.
- = - = - - = - = - - = - = - - = - = -
(INFO/examples:)
cat /etc/sysconfig/network-scripts/ifcfg-eth0 cat /etc/resolv.conf
PBX in a Flash PURPLE Daemon Status - Version 1.8.1 Released on 041211 System Information Asterisk = ONLINE | Dahdi = ONLINE | MySQL = ONLINE SSH = ONLINE | Apache = ONLINE | Iptables = ONLINE Fail2ban = ONLINE | Internet = OFFLINE | Ip6Tables = ONLINE BlueTooth = ONLINE | Hidd = ONLINE | NTPD = ONLINE SendMail = ONLINE | Samba = OFFLINE | Webmin = ONLINE Ethernet0 = ONLINE | Ethernet1 = N/A | Wlan0 = N/A
PBX in a Flash Version = 1.7.5.6 FreePBX Version = 2.8.1.4 Running Asterisk Version = 1.8.3.3 Asterisk Source Version = 1.8.3.3 Dahdi Source Version = 2.4.1.2+2.4.1 Libpri Source Version = 1.4.11.5 IP Address = 192.168.70.40 on eth0 Operating System = CentOS release 5.6 (Final) Kernel Version = 2.6.18-238.9.1.el5 - 32 Bit
root@pbx:/etc/sysconfig/network-scripts $ cat ifcfg-eth0 # Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ DEVICE=eth0 TYPE=ethernet BOOTPROTO=none BROADCAST=192.168.70.255 IPADDR=192.168.70.40 NETMASK=255.255.255.0 GATEWAY=192.168.70.2 HWADDR=00:1D:7D:26:31:D0 ONBOOT=yes DHCP_HOSTNAME=pbx.local
root@pbx:/etc/sysconfig/network-scripts $
root@pbx:/etc $ cat resolv.conf search isp.com nameserver 216.146.35.35 nameserver 216.146.36.36 nameserver 8.8.8.8
root@pbx:/etc $ ping www.henrichs.org ping: unknown host www.henrichs.org
root@pbx:/etc $ ping 69.64.155.165 PING 69.64.155.165 (69.64.155.165) 56(84) bytes of data. 64 bytes from 69.64.155.165: icmp_seq=1 ttl=238 time=92.9 ms 64 bytes from 69.64.155.165: icmp_seq=2 ttl=238 time=89.1 ms 64 bytes from 69.64.155.165: icmp_seq=3 ttl=238 time=89.1 ms 64 bytes from 69.64.155.165: icmp_seq=4 ttl=238 time=89.7 ms 64 bytes from 69.64.155.165: icmp_seq=5 ttl=238 time=88.3 ms 64 bytes from 69.64.155.165: icmp_seq=6 ttl=238 time=90.6 ms
--- 69.64.155.165 ping statistics --- 6 packets transmitted, 6 received, 0% packet loss, time 5002ms rtt min/avg/max/mdev = 88.398/90.013/92.994/1.514 ms
root@pbx:/etc $ tracert www.henrichs.org www.henrichs.org: Temporary failure in name resolution Cannot handle "host" cmdline arg `www.henrichs.org' on position 1 (argc 1)
root@pbx:/etc $ dig www.henrichs.org
; <<>> DiG 9.3.6-P1-RedHat-9.3.6-16.P1.el5 <<>> www.henrichs.org ;; global options: printcmd ;; Got answer: ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 48631 ;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 5, ADDITIONAL: 0
;; QUESTION SECTION: ;www.henrichs.org. IN A
;; ANSWER SECTION: www.henrichs.org. 3566 IN CNAME ehost-services119.com. ehost-services119.com. 3565 IN A 69.64.155.165
;; AUTHORITY SECTION: ehost-services119.com. 3565 IN NS dns1.name-services.com. ehost-services119.com. 3565 IN NS dns2.name-services.com. ehost-services119.com. 3565 IN NS dns3.name-services.com. ehost-services119.com. 3565 IN NS dns4.name-services.com. ehost-services119.com. 3565 IN NS dns5.name-services.com.
;; Query time: 18 msec ;; SERVER: 216.146.35.35#53(216.146.35.35) ;; WHEN: Wed Jun 22 12:17:53 2011 ;; MSG SIZE rcvd: 194
root@pbx:/etc $ wget www.henrichs.org --2011-06-22 12:18:38-- http://www.henrichs.org/ Resolving www.henrichs.org... failed: Temporary failure in name resolution. wget: unable to resolve host address `www.henrichs.org' root@pbx:/etc $ nslookup www.henrichs.org Server: 216.146.35.35 Address: 216.146.35.35#53
Non-authoritative answer: www.henrichs.org canonical name = ehost-services119.com. Name: ehost-services119.com Address: 69.64.155.165
root@pbx:/etc $ route Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface 192.168.70.0 * 255.255.255.0 U 0 0 0 eth0 169.254.0.0 * 255.255.0.0 U 0 0 0 eth0 default 192.168.70.2 0.0.0.0 UG 0 0 0 eth0 root@pbx:/etc $
-----Original Message----- Behalf Of robert henrichs Subject: [CentOS] inconsistent DNS results - ping vs dig vs nslookup
I am getting what I believe to be inconsistent replies/results from my CentOs machine, trying to ping URLs. Pinging by URL fails the DNS lookup, while pinging the IP address (that should have been returned) works. The DNS lookup also fails NSLOOKUP, but interestingly, dig and wget return A records with the correct IP.
What does your "hosts:" entry in /etc/nsswitch.conf look like? Does is list dns as a source?
Mine looks like this:
hosts: files dns
-- Owen Beckley
robert henrichs wrote:
I am getting what I believe to be inconsistent replies/results from my CentOs machine, trying to ping URLs. Pinging by URL fails the DNS lookup, while pinging the IP address (that should have been returned) works. The DNS lookup also fails NSLOOKUP, but interestingly, dig and wget return A records with the correct IP.
Robert Henrichs.
wget has *not* returned correct IP! Only nslookup and dig.
First try to use only google DNS servers: 8.8.8.8 and 8.8.4.4, and if that does not help, and your only problem is with only that domain, you will have to contact support where domain is parked, and see is that helps.
Only relevant info is in resolve.conf, there is no need to look at network configuration if you can ping IP's.
Ljubomir
Thanks All,
Answering multiple questions:
*** I started off with Google's DNS servers- 8.8.8.8 and 8.8.4.4 with the same results. The current set were copied off other working wkstas on the LAN . Same type results with "ping www.yahoo.com" and "ping www.yahoo.com". Just changed it to: (1) 8.8.4.4 (2) 216.146.36.36 (3) 8.8.8.8
Rebooted the box, then . . .
root@pbx:~ $ cat /etc/resolv.conf search isp.com nameserver 8.8.4.4 nameserver 216.146.36.36 nameserver 8.8.8.8 root@pbx:~ $ ping www.henrichs.org ping: unknown host www.henrichs.org root@pbx:~ $ ping www.yahoo.com ping: unknown host www.yahoo.com root@pbx:~ $ ping yahoo.com ping: unknown host yahoo.com root@pbx:~ $ ping microsoft.com ping: unknown host microsoft.com root@pbx:~ $ ping lilug.org ping: unknown host lilug.org root@pbx:~ $ ping centos.org ping: unknown host centos.org root@pbx:~ $
what's a good link to read about how the resolv.conf file is constructed? esp. the "search" entry?
*** This was part of the wget response: www.henrichs.org canonical name = ehost-services119.com. Name: ehost-services119.com Address: 69.64.155.165
Doesn't the "address: Address: 69.64.155.165" indicate a correct result?
*** My nsswitch.conf file does contain: hosts: files dns (this is an interesting file I'll have to revisit...thanks).
***
It was suggested off-list that I show "the contents of /etc/resolv.conf and a little history on your ifcfg-ethx files in /etc/sysconfig/network-scripts/"
Here is /etc/resolv.conf: root@pbx:/etc/sysconfig/network-scripts $ cat /etc/resolv.conf search isp.com nameserver 216.146.35.35 nameserver 216.146.36.36 nameserver 8.8.8.8 root@pbx:/etc/sysconfig/network-scripts $
Here are the files in my /etc/sysconfig/network-scripts directory: root@pbx:/etc/sysconfig/network-scripts $ ls ifcfg-eth0 ifdown-ippp ifdown-ppp ifup ifup-ib ifup-isdn ifup-routes init.ipv6-global ifcfg-lo ifdown-ipsec ifdown-routes ifup-aliases ifup-ippp ifup-plip ifup-sit net.hotplug ifdown ifdown-ipv6 ifdown-sit ifup-bnep ifup-ipsec ifup-plusb ifup-sl network-functions ifdown-bnep ifdown-isdn ifdown-sl ifup-eth ifup-ipv6 ifup-post ifup-tunnel network-functions-ipv6 ifdown-eth ifdown-post ifdown-tunnel ifup-hdlc ifup-ipx ifup-ppp ifup-wireless root@pbx:/etc/sysconfig/network-scripts $
Here is the contents of the ifcfg-eth0 file: root@pbx:/etc/sysconfig/network-scripts $ cat ifcfg-eth0 # Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ DEVICE=eth0 TYPE=ethernet BOOTPROTO=none BROADCAST=192.168.70.255 IPADDR=192.168.70.40 NETMASK=255.255.255.0 GATEWAY=192.168.70.2 HWADDR=00:1D:7D:26:31:D0 ONBOOT=yes DHCP_HOSTNAME=pbx.local root@pbx:/etc/sysconfig/network-scripts $
Robert Henrichs.
At 12:42 PM 6/22/2011, you wrote:
robert henrichs wrote:
I am getting what I believe to be inconsistent replies/results from my CentOs machine, trying to ping URLs. Pinging by URL fails the DNS lookup, while pinging the IP address (that should have been returned) works. The DNS lookup also fails NSLOOKUP, but interestingly, dig and wget return A records with the correct IP.
Robert Henrichs.
wget has *not* returned correct IP! Only nslookup and dig.
First try to use only google DNS servers: 8.8.8.8 and 8.8.4.4, and if that does not help, and your only problem is with only that domain, you will have to contact support where domain is parked, and see is that helps.
Only relevant info is in resolve.conf, there is no need to look at network configuration if you can ping IP's.
Ljubomir _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
what's a good link to read about how the resolv.conf file is constructed? esp. the "search" entry?
man resolv.conf
This was part of the wget response: www.henrichs.org canonical name = ehost-services119.com. Name: ehost-services119.com Address: 69.64.155.165
Doesn't the "address: Address: 69.64.155.165" indicate a correct result?
That's the response from nslookup, not wget.
How is this server connected to the Internet? Sounds like something is blocking domain port 53.
On 6/22/2011 3:30 PM, Robert Henrichs wrote:
what's a good link to read about how the resolv.conf file isconstructed? esp. the "search" entry?
The 'search' should only apply if you don't supply the domain portion of the name. Each nameserver address should be queried until one responds.
My nsswitch.conf file does contain: hosts: files dns (this is an interesting file I'll have to revisit...thanks).
The part that doesn't make sense yet is that your dns is working as shown by the dig/nslookup queries, but ping/wget are failing to look up the name. You might get random success/failure if some of the nameservers in resolv.conf are misconfigured. Seems like a long shot but try: dig @server_address www.henrichs.org for each of the namesever addresses, then maybe strace ping www.henrichs.org and wade through the output to see if it has trouble loading the resolver library or something.
Robert Henrichs wrote:
root@pbx:~ $ cat /etc/resolv.conf search isp.com nameserver 8.8.4.4 nameserver 216.146.36.36 nameserver 8.8.8.8
Whats with the "search isp.com"? Try removing that and try again. Try first without rebooting.
I am looking what it actually does. I do not have it on my systems.
Ljubomir
Ljubomir Ljubojevic wrote:
Robert Henrichs wrote:
root@pbx:~ $ cat /etc/resolv.conf search isp.com nameserver 8.8.4.4 nameserver 216.146.36.36 nameserver 8.8.8.8
Whats with the "search isp.com"? Try removing that and try again. Try first without rebooting.
I am looking what it actually does. I do not have it on my systems.
Ljubomir
As Les wrote in another branch of the thread, search clause is if you try name without a domain.
Anyway, try removing it, but first try using ONLY 8.8.8.8 and 8.8.4.4, remove 216.146.36.36 and then test.
Ljubomir
On Wed, 22 Jun 2011, Ljubomir Ljubojevic wrote:
As Les wrote in another branch of the thread, search clause is if you try name without a domain.
I think it's slightly more subtle and possibly more annoying than that.
Say you have a machine called foo.mydomain.com
By default (if you don't specify it), search is set to be mydomain.com.
So a lookup for bar would initially be done as bar.domain.com.
A lookup for bar.baz would initially be searched for as bar.baz (because it has at least ndots in it), but if that failed, would then be looked up as bar.baz.domain.com.
Equally a search for baz.domain.com (where baz didn't exist) would be looked up first as baz.domain.com and then as baz.domain.com.domain.com.
This appears to be doubly annoying with programs compile with IPv6 support where there aren't AAAA records.
So a typical valid lookup for machine.domain.com becomes:
AAAA lookup for machine.domain.com (not found) AAAA lookup for machine.domain.com.domain.com (not found) A lookup for machine.domain.com (found)
jh
Can you avoid the bar.baz.domain.com.domain.com by searching for
bar.baz.domain.com.
(note trailing dot)
??
On Thu, June 23, 2011 11:06, John Hodrien wrote:
On Wed, 22 Jun 2011, Ljubomir Ljubojevic wrote:
As Les wrote in another branch of the thread, search clause is if you try name without a domain.
I think it's slightly more subtle and possibly more annoying than that.
Say you have a machine called foo.mydomain.com
By default (if you don't specify it), search is set to be mydomain.com.
So a lookup for bar would initially be done as bar.domain.com.
A lookup for bar.baz would initially be searched for as bar.baz (because it has at least ndots in it), but if that failed, would then be looked up as bar.baz.domain.com.
Equally a search for baz.domain.com (where baz didn't exist) would be looked up first as baz.domain.com and then as baz.domain.com.domain.com.
This appears to be doubly annoying with programs compile with IPv6 support where there aren't AAAA records.
So a typical valid lookup for machine.domain.com becomes:
AAAA lookup for machine.domain.com (not found) AAAA lookup for machine.domain.com.domain.com (not found) A lookup for machine.domain.com (found)
jh _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On Thu, 23 Jun 2011, Giles Coochey wrote:
Can you avoid the bar.baz.domain.com.domain.com by searching for
bar.baz.domain.com.
(note trailing dot)
??
Hmm, good suggestion, that I'd not considered, Thanks. It does appear to clear that up (down to two lookups from three: an AAAA and an A). It doesn't always appear to work though:
http://bar.baz.domain.com./index.html
This doesn't appear to universally work, but I'm not sure whether that's a bug with the affected web servers or not.
jh
On Thu, June 23, 2011 11:53, John Hodrien wrote:
On Thu, 23 Jun 2011, Giles Coochey wrote:
Can you avoid the bar.baz.domain.com.domain.com by searching for
bar.baz.domain.com.
(note trailing dot)
??
Hmm, good suggestion, that I'd not considered, Thanks. It does appear to clear that up (down to two lookups from three: an AAAA and an A). It doesn't always appear to work though:
http://bar.baz.domain.com./index.html
This doesn't appear to universally work, but I'm not sure whether that's a bug with the affected web servers or not.
Yes, I'm sure it will depend on the implementation, the trailing dot was somewhat an educated guess from previous ISC BIND & dig tool use. :-) As for both the A and AAAA record, I think you will have that until IPv4 is fully deprecated by IPv6 and no longer exists, unless, of course, you want to use non IPv6 aware software (or non-IPv4 aware software... shudder - it will happen!).
On Thu, 23 Jun 2011, Giles Coochey wrote:
Yes, I'm sure it will depend on the implementation, the trailing dot was somewhat an educated guess from previous ISC BIND & dig tool use. :-) As for both the A and AAAA record, I think you will have that until IPv4 is fully deprecated by IPv6 and no longer exists, unless, of course, you want to use non IPv6 aware software (or non-IPv4 aware software... shudder
- it will happen!).
I think software still could make a better stab at things now. You're doing AAAA lookups on a system without any IPv6 interfaces. It's not like it's useful information to have if it succeeds...
jh
On Thu, June 23, 2011 12:07, John Hodrien wrote:
On Thu, 23 Jun 2011, Giles Coochey wrote:
Yes, I'm sure it will depend on the implementation, the trailing dot was somewhat an educated guess from previous ISC BIND & dig tool use. :-) As for both the A and AAAA record, I think you will have that until IPv4 is fully deprecated by IPv6 and no longer exists, unless, of course, you want to use non IPv6 aware software (or non-IPv4 aware software... shudder
- it will happen!).
I think software still could make a better stab at things now. You're doing AAAA lookups on a system without any IPv6 interfaces. It's not like it's useful information to have if it succeeds...
Well, if you have access to the source, you can probably modify some constants to disable particular protocols, or there may be compile time options already - it would be nice to disable automatically on runtime if the host does not have a IPv6 protocol stack.
The system calls in question are:
getaddrinfo(3) http://www.kernel.org/doc/man-pages/online/pages/man3/getaddrinfo.3.html
or in older software:
gethostbyname(3) http://www.kernel.org/doc/man-pages/online/pages/man3/gethostbyname.3.html
On Thu, 23 Jun 2011, Giles Coochey wrote:
Well, if you have access to the source, you can probably modify some constants to disable particular protocols, or there may be compile time options already - it would be nice to disable automatically on runtime if the host does not have a IPv6 protocol stack.
The system calls in question are:
getaddrinfo(3) http://www.kernel.org/doc/man-pages/online/pages/man3/getaddrinfo.3.html
or in older software:
gethostbyname(3) http://www.kernel.org/doc/man-pages/online/pages/man3/gethostbyname.3.html
My understanding was that AI_ADDRCONFIG does exactly that, only doing a given lookup if IPv6 is configured. If that's the case, then it's just woefully underused.
jh
--On Wednesday, June 22, 2011 03:30:18 PM -0500 Robert Henrichs centos@henrichs.biz wrote:
root@pbx:~ $ cat /etc/resolv.conf search isp.com nameserver 8.8.4.4 nameserver 216.146.36.36 nameserver 8.8.8.8
Get rid of the leading whitespace if it actually exists in that file. It shouldn't make a difference, but who knows.
BTW, you don't need to reboot.
root@pbx:~ $ ping www.yahoo.com ping: unknown host www.yahoo.com
Try a trailing dot on the FQDN and see if it makes a difference. It shouldn't with ping, but should always be used with dig:
ping www.yahoo.com.
The trailing dot changes the lookup algorithm a bit.
You could also do a RES_OPTIONS="debug" export RES_OPTIONS ping www.yahoo.com. to try to figure things out.
Devin
Kevin Thorpe wrote:
> Hi guys, any CUPS / Samba experts out there?
Bah! I saw that lpr could print PS but smbclient coudn't so in desperation I restarted samba. Now it prints PS. Why is completely beyond me. That's why I couldn't work out what I'd done wrong - I hadn't.
Following up my last post, I just looked at the smbd manpage - this may be of interest, from the very *bottom* of the manpage:
<snip> SIGNALS
Sending the smbd a SIGHUP will cause it to reload its smb.conf configu- ration file within a short period of time. <snip>
mark
On Wed, Jun 22, 2011 at 5:41 PM, m.roth@5-cent.us wrote:
Kevin Thorpe wrote:
>> Hi guys, any CUPS / Samba experts out there?
Bah! I saw that lpr could print PS but smbclient coudn't so in desperation I restarted samba. Now it prints PS. Why is completely beyond me. That's why I couldn't work out what I'd done wrong - I hadn't.
Following up my last post, I just looked at the smbd manpage - this may be of interest, from the very *bottom* of the manpage:
<snip> SIGNALS
Sending the smbd a SIGHUP will cause it to reload its smb.conf configu- ration file within a short period of time.
Yup, I tried SIGHUP. That didn't fix it but restart did - I guess Samba is a bit too much like windows ;-)
At Wed, 22 Jun 2011 16:22:57 +0100 CentOS mailing list centos@centos.org wrote:
On Wed, 22 Jun 2011 15:16:54 +0100 Kevin Thorpe kevin.thorpe@pibenchmark.com wrote:
Hi guys, any CUPS / Samba experts out there?
In CUPS https://localhost:631/ setup your printers and set the printer driver for each printer to local raw printer
And you say that, using localhost:631, you did add printer, then selected make and model, and chose one of the drivers, and it still doesn't do the right thing?
That sets the printer up just fine in Linux. I can print to that fine with Open Office. The problem I have is that a generic PS driver in Windows to the SMB printer outputs the raw PS script instead of rendering which is odd because the CUPS test page is a postscript file.
Look *closely* at your smb.conf file. I'll bet it has something like this:
# Cups Options let you pass the cups libs custom options, setting it to raw # for example will let you use drivers on your Windows clients # # Printcap Name let you specify an alternative printcap file # # You can choose a non default printing system using the Printing option
load printers = yes cups options = raw
; printcap name = /etc/printcap #obtain list of printers automatically on SystemV printcap name = cups show add printer wizard = no printing = cups
Note the line 'cups options = raw'. This means that when you add a printer via NETBIOS (adding a network printer from the Network Neighborhood) and send a printout, it gets sent *directly* to the printer and DOES NOT go through the CUPS (foomatic) filter. This means that the MS-Windows machine must have the MS-Windows printer driver for the printer installed and must be using this driver.
For a *real* PostScript printer (eg almost all *Laser* printers (typically all but bargin basement types)), picking the 'generic' PostScript driver will work just fine (I have no real idea *why* MS-Windows thinks you need to have umpteen *different* PostScript drivers, one for each model of PostScript printer).
MS-Windows *supposedly* supports the ipp protocol. If you manage to get the MS-Windows box to use *that* protocol for the network printer, (eg ipp://IP-address:631/...), then you *should* be able to use the 'generic' PostScript, even if the printer is your basic (cheap) inkjet printer using the CUPS foomagic (Ghostscript) driver/filter.
-
Benjamin Hackl -
Devin Reade -
Giles Coochey -
John Hodrien -
Kevin Thorpe -
Les Mikesell -
Ljubomir Ljubojevic -
m.roth＠5-cent.us -
Owen Beckley -
Robert Heller -
robert henrichs -
Ryan Ivey