Topic: SAIL Installation on SME 7.1

[ntblade]

Hi all,
Sorry to be a pest, Selintra but I'm still having problems installing SAIL so, here are the steps I'm taking (fresh 7.1 installation)...

Code: [Select]

[root@test SAIL]# uname -r
2.6.9-42.0.3.EL
[root@test SAIL]# rpm -Uvh smeserver-asterisk-zappri-MPP-1.2.6-1.i686.rpm
Preparing...                ########################################### [100%]
   1:smeserver-asterisk-zapp########################################### [100%]
[root@test SAIL]# rpm -Uvh smeserver-asterisk-1.2.10-3.i686.rpm
Preparing...                ########################################### [100%]
   1:smeserver-asterisk     ########################################### [100%]
Notice: Configuration file is /etc/zaptel.conf
line 0: Unable to open master device '/dev/zap/ctl'

1 error(s) detectedI'm sure that "Shutting down Asterisk" appears and is over-written by the "1 error(s) detected" line.
Is this normal?

Thanks
Norrie

[hervep]

Hi ntblade,

Here is how I proceed to install 'Sail' onto 'fresh SME 7.1'.
After reboot ... it works.

Rpm's used :

nmap-3.70-1.i386.rpm
selintra-sail-2.1.14-425.noarch.rpm
smeserver-asterisk-1.2.10-3.i686.rpm
smeserver-asterisk-sounds-1.2.2-2.noarch.rpm
smeserver-asterisk-zappri-MPP-1.2.6-1.i686.rpm
tftp-server-0.39-2.i386.rpm
xinetd-2.3.13-4.4E.1.i386.rpm

Commands :

rpm -Uvh *.rpm
/sbin/e-smith/signal-event post-upgrade
/sbin/e-smith/db configuration setprop yum status disabled
/sbin/e-smith/db configuration setprop xinetd status enabled
/sbin/e-smith/db configuration setprop tftp status enabled
/sbin/e-smith/signal-event reboot

Hope this can help ...

Hervé

[del]

Hi hervep,

I don't have this installed:

nmap-3.70-1.i386.rpm

What is it for and do I need it My SAIL PBX seems to work OK. Where is the best place to download these rpms:

tftp-server-0.39-2.i386.rpm
xinetd-2.3.13-4.4E.1.i386.rpm

I have them installed but they are older versions. Thanks for your help.

Regards,
Del

[ntblade]

Thanks for the replies,

I'm trying to get the install to work as per the instructions here...
http://www.selintra.com/docs/cgi-bin/view/Main/DocChapter03

If you DON'T wish to run ISDN BRI cards then install the rpms at

http://mirror.contribs.org/smeserver//contribs/selintra/RPMS/AsteriskForSail-2.1.13/None-ISDN/

Finally, install selintra-sail-2.1.14-nnn which you will find here...

ftp://81.149.154.14/Pre-Releases/

All pre -389 rpms should be installed as follows...

rpm -Uvh target-rpm-release

NB - SAIL-389 onwards

Sail -389 requires nmap so you should install it with yum as follows....

yum localinstall selintra-sail-2.1.14-389.noarch.rpm --enablerepo=base

Cleanup

Once all of your rpms are installed, you must run console-save...

/sbin/e-smith/signal-event console-save

Extras

If you have not yet done it and you want to run automatic provisioning via the on-board Sail tftp-server then you should install tftp-server

Enable CentOS base and install tftp-server as follows...

yum install tftp-server --enablerepo=base

So, I guess this is as mutch about testing the documentation as well as tesing the software but I'm not 100% confident that my installation is as expected as halfway through I get:

Code: [Select]

WARNING: Template processing succeeded for /: 8 fragments generated warnings
 at /usr/lib/perl5/site_perl/esmith/DB.pm line 459
Migrating existing database yum_updates
Migrating existing database configuration
Migrating existing database networks
Migrating existing database accounts
Migrating existing database backups
Migrating existing database spamassassin
Use of uninitialized value in substitution (s///) at /usr/lib/perl5/site_perl/esmith/FormMagick.pm line 237, <LEXIN> line 1.
Use of uninitialized value in subroutine entry at /usr/lib/perl5/vendor_perl/5.8.5/i386-linux-thread-multi/XML/Parser/Expat.pm line 474.

<snip>

Use of uninitialized value in print at /usr/lib/perl5/site_perl/CGI/FormMagick/Setup.pm line 179.
smeserver trap post-install: selintra-sail
smeserver trap post-install: nmap

Installed: selintra-sail.noarch 0:2.1.14-425
Dependency Installed: nmap.i386 2:3.70-1
Complete!
==============================================================
WARNING: You now need to run BOTH of the following commands
to ensure consistent system state:

signal-event post-upgrade; signal-event reboot

You should run these commands unless you are certain that
yum made no changes to your system.
==============================================================
Why are you installing xinetd?

After the reboot I go to PCI Cards (none installed at the moment) press probe and get

Code: [Select]

Operation status report - Timed out on UDEV - It may work if you retry...
but the ztdummy module should then be installed

Thanks

[del]

Hi All,

Please excuse my stupidity

I don't have this installed:
Quote:
nmap-3.70-1.i386.rpm
What is it for and do I need it Shocked My SAIL PBX seems to work OK.

I am still using selintra-sail-2.1.14-339.noarch.rpm, which explains why everything is running OK

Regards,
Del

[hervep]

Hi hervep,

I don't have this installed:

nmap-3.70-1.i386.rpm

What is it for and do I need it My SAIL PBX seems to work OK. Where is the best place to download these rpms:

nmap-3.70-1.i386.rpm
tftp-server-0.39-2.i386.rpm
xinetd-2.3.13-4.4E.1.i386.rpm

I have them installed but they are older versions. Thanks for your help.

Regards,
Del

nmap is required for the automatic sets discovery.
It is a 'network scanner'. It may be not needed in you use older version of 'Sail'.

You can use copy-paste in google to find them, or rpmfind.net ... .
If your actual version works, no problem. Those are not critical 'sail' rpm's.

Kind regards,

Hervé

[hervep]

Thanks for the replies,

I'm trying to get the install to work as per the instructions here...
http://www.selintra.com/docs/cgi-bin/view/Main/DocChapter03

If you DON'T wish to run ISDN BRI cards then install the rpms at

http://mirror.contribs.org/smeserver//contribs/selintra/RPMS/AsteriskForSail-2.1.13/None-ISDN/

Finally, install selintra-sail-2.1.14-nnn which you will find here...

ftp://81.149.154.14/Pre-Releases/

All pre -389 rpms should be installed as follows...

rpm -Uvh target-rpm-release

NB - SAIL-389 onwards

Sail -389 requires nmap so you should install it with yum as follows....

yum localinstall selintra-sail-2.1.14-389.noarch.rpm --enablerepo=base

Cleanup

Once all of your rpms are installed, you must run console-save...

/sbin/e-smith/signal-event console-save

Extras

If you have not yet done it and you want to run automatic provisioning via the on-board Sail tftp-server then you should install tftp-server

Enable CentOS base and install tftp-server as follows...

yum install tftp-server --enablerepo=base

So, I guess this is as mutch about testing the documentation as well as tesing the software but I'm not 100% confident that my installation is as expected as halfway through I get:

Code: [Select]

WARNING: Template processing succeeded for /: 8 fragments generated warnings
 at /usr/lib/perl5/site_perl/esmith/DB.pm line 459
Migrating existing database yum_updates
Migrating existing database configuration
Migrating existing database networks
Migrating existing database accounts
Migrating existing database backups
Migrating existing database spamassassin
Use of uninitialized value in substitution (s///) at /usr/lib/perl5/site_perl/esmith/FormMagick.pm line 237, <LEXIN> line 1.
Use of uninitialized value in subroutine entry at /usr/lib/perl5/vendor_perl/5.8.5/i386-linux-thread-multi/XML/Parser/Expat.pm line 474.

<snip>

Use of uninitialized value in print at /usr/lib/perl5/site_perl/CGI/FormMagick/Setup.pm line 179.
smeserver trap post-install: selintra-sail
smeserver trap post-install: nmap

Installed: selintra-sail.noarch 0:2.1.14-425
Dependency Installed: nmap.i386 2:3.70-1
Complete!
==============================================================
WARNING: You now need to run BOTH of the following commands
to ensure consistent system state:

signal-event post-upgrade; signal-event reboot

You should run these commands unless you are certain that
yum made no changes to your system.
==============================================================
Why are you installing xinetd?

After the reboot I go to PCI Cards (none installed at the moment) press probe and get

Code: [Select]

Operation status report - Timed out on UDEV - It may work if you retry...
but the ztdummy module should then be installed

Thanks

Looks great ...

Xinetd is needed for tftp.

Only difference is that I do it as a 'local' installation ( RPM localy available ).
It does not requires internet/yum connection during installation.

Edit : Ztdummy depends also of the hardware you have on the pc.
         It is a 'virtual' clock source that is needed for some applications as
         'conferencing'. If it don't start, it is not a critical issue.

Best,

Herve

[ntblade]

[/quote]Looks great ...

Xinetd is needed for tftp.

Only difference is that I do it as a 'local' installation ( RPM localy available ).
It does not requires internet/yum connection during installation.

Lokks fine but asterisk doesn't start on boot???

[hervep]

Lokks fine but asterisk doesn't start on boot???

I do not have the same problem, difficult to answer ...
Assuming you can start asterisk manually, I would suggest to check the following :

- Did you performed ' signal-event post-upgrade & signal-event reboot  ' ?
- Check http://www.selintra.com/docs/cgi-bin/view/Main/DocChapter2510
- Check if service start Symlink is OK. ( can you find 'S93asterisk' into /etc/rc.d/rc7.d ? ).  If not :

Code: [Select]

#/bin/ln -s /etc/rc.d/init.d/e-smith-service /etc/rc.d/rc7.d/S93asterisk

Best,

Hervé

[SARK devs]

Hi guys

You all look to be doing just fine...

However.....

I'm sure that "Shutting down Asterisk" appears and is over-written by the "1 error(s) detected" line.
Is this normal?

Yes, I'm afraid it is.  This has to do with the strange way in which rpm works and our ignorance when we rolled the original asterisk rpms.  The good news is that it isn't critical


Now, here's the thing...

I ran a clean install of SAIL-425 on SME 7.1 this evening (this is something which we very rarely do because we normally use our SARK.iso to install the lot in one go) and I ran into a couple of problems.

I effectively needed to run the install twice because of the kernel differences.  So...

I will update the zaptel rpms over the weekend to make life easier for everyone.  However, in the meantime here is what I had to do to get a clean SAIL install on 7.1....

drop the rpms (smeserver-asterisk-zappri-MPP, smeserver-asterisk, lastest version of sail (-425) and whichever language modules you want)

Step 1 - install zaptel....

rpm -Uvh  smeserver-asterisk-zappri-MPP-whatever

Now copy the "extra" subdirectory into the correct kernel tree.

e.g. cp -r /lib/modules/2.6.9-34.EL/extra /lib/modules/2.6.9-42.03.EL/

Now remove zaptel...

rpm -e smeserver-asterisk-zappri-MPP

and reinstall it....

rpm -Uvh smeserver-asterisk-zappri-MPP-whatever

I know, I know, it's a complete faff but it's what you need to do right now  
The rest of the install is as described in the docs.

Promise we'll upgrade the zaptel rpms at the weekend if we get time.



J

[ntblade]

Hi Selintra,

Did all that, rebooted and asterisk doesn't start.
Went to PCI cards, Probe and:
"Operation status report - Timed out on UDEV - It may work if you retry..."
Pressed "Start" and then "Probe" and the dummy module seems to load.
Went to globals and filled in IP address mail etc, save, commit, rebooted and asterisk still doesn't start but pressing "Start" at the PCI Cards panel starts asterisk and the zaptel module loads...

Code: [Select]

[root@test ~]# lsmod
Module                  Size  Used by
ztdummy                 3924  0
zaptel                206852  3 ztdummy
crc_ccitt               2241  1 zaptel
Any ideas?

Thanks
N

[SARK devs]

HI NT

Yes, I had the same issue.  The link is not being created properly. Do...

Code: [Select]

/bin/ln -s /etc/rc.d/init.d/e-smith-service /etc/rc.d/rc7.d/S93asterisk

I think that will solve the start-up problem.

Other than that you look good to go.



J

[ntblade]

Thanks!

Installation and startup problems now 99% sorted.

On satrtup I'm getting...

Code: [Select]

Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 50: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 51: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 52: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 53: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 54: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 55: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 56: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 57: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 58: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 59: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 60: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 61: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 62: ignoring bad line starting with 'post-install'
Mar  4 11:57:55 bacon modprobe: WARNING: /etc/modprobe.conf line 63: ignoring bad line starting with 'post-install'

but this doesn't seem to be a problem.

Also, I can't see anywhere in the manual f it's possible to automatically prepend our local code (not using PSTN at the the moment - Voiptalk only) so that all calls beginning with 8 will go out as 016208xxxxx say.


Edit:

Here's my Transformation Mask...
00: 0:44 8:4416208
and Dial Plan...
_0. _8XXXXX

Seems to work but what else should go in?

All the best

Norrie

[SARK devs]

Hi Norrie,

The messages are coming out of the kernel.  This is due to the fact that it can't find the zaptel modules in the correct module tree.  You have to move the extras subdirectory to the new kernel tree -  see here...

http://www.selintra.com/docs/cgi-bin/view/Main/DocChapter2510

Once you've moved 'em you then need to remove and reinstall zaptel (as per my previous post).  Unfortunately, because of the dependencies you need to remove sail and asterisk also and then re-install everything.  If you don't do this then you will not be able to run any digium cards and things like conferencing (which need a timing source) wont work.

We'll get around to fixing this snafu soon but for the moment this above is the workaround.

re area codes...

You can prepend area codes in the trunk mask.  It's not very elegant because it's not really what masks were designed for but it works fine.  You need to know the start digits of your local numbers .  In the UK there will usually be two, the BT start digit and the NTL(Mercury) start digit.  So, for example, in our area, subscriber numbers can start with either a 4 or 5.  Our area code is 01924 so you can create a mask like this...

Code: [Select]

4:019244 5:019245

If you aren't sure what the local start digits are then you can just do 'em all like this...

Code: [Select]

2:019242 3:019243 4:019244 5:019245 6:019246 7:019247 8:019248 9:019249

Job done.


Kind Regards

Jeff[/code]

[ntblade]

Thanks for the reply, Jeff - on a Sunday too!

I did a bit of digging and found the transformation mask thread and looked a bit more in the manual. and put:
00: 0:44 8:4416208
in my mask.  This works but:
00: 0:44 8:016208
doesn't.

I followed your revised installtion instructions:
Installed zaptel
copied the extras directory:

Code: [Select]

ls -l /lib/modules/2.6.9-42.0.3.EL/extra/
total 4324
-rw-r--r--  1 root root  254728 Mar  3 16:34 pciradio.ko
-rw-r--r--  1 root root  219335 Mar  3 16:34 tor2.ko
-rw-r--r--  1 root root  156530 Mar  3 16:34 torisa.ko
-rw-r--r--  1 root root  138569 Mar  3 16:34 wcfxo.ko
-rw-r--r--  1 root root  206881 Mar  3 16:34 wct1xxp.ko
-rw-r--r--  1 root root 1372940 Mar  3 16:34 wct4xxp.ko
-rw-r--r--  1 root root  413461 Mar  3 16:34 wctdm24xxp.ko
-rw-r--r--  1 root root  214484 Mar  3 16:34 wctdm.ko
-rw-r--r--  1 root root  243431 Mar  3 16:34 wcte11xp.ko
-rw-r--r--  1 root root  173856 Mar  3 16:34 wcusb.ko
-rw-r--r--  1 root root  465454 Mar  3 16:34 zaptel.ko
-rw-r--r--  1 root root  132692 Mar  3 16:34 ztd-eth.ko
-rw-r--r--  1 root root  116260 Mar  3 16:34 ztd-loc.ko
-rw-r--r--  1 root root   93022 Mar  3 16:34 ztdummy.ko
-rw-r--r--  1 root root  131965 Mar  3 16:34 ztdynamic.ko

uninstalled and reinstalled zaptel then installed all of the rest of the apps.
Did a depmod, signal-event post-upgrade reboot etc.
The dummy module sems to have been loaded fine:

lsmod
Module                  Size  Used by
ztdummy                 3924  0
zaptel                206852  3 ztdummy
crc_ccitt               2241  1 zaptel

So do I really need to uninstall and reinstall all the rest of SAIL?

Also (and you might throw your hands up in the air) I want to run this server in a DMZ behind an IPCop firewall.  This installation is located where the ADSL is shared between two builings on different LANs.  both should have access the the server but one is also in the DMZ.  There's to be a separate wireless LAN which will be used to bridge to yet another building which will have internet access only. - no server access

Here's a diagram...

Code: [Select]

Wireless LAN - Phones later?
|          |         |
+----------+---------+
192.168.2.0|
           |         DMZ LAN - No Phones but other client PCs
           |         +----------------
       +---+---+     |       +-------+
  ADSL |       | DMZ |       |       |
 ------+ IPCop +-----+-------+  SME  |
       |       | 192.168.1.0 |       |
       +---+---+             +-------+
           |
192.168.0.0|
 +---------+---------+
 |         |         |
Private LAN - Phones here

So
PCs in private LAN can access DMZ but PCs in DMZ can't access Private LAN but can access Internet
PCs in wireless LAN can access Internet and DMZ but not Private LAN

Phones in DMZ initially register but are then "unreachable" and can call phones in private LAN but there is no sound.
Phones in Private LAN cannot call phones in DMZ but can do everything else.

I really want to keep the server in the DMZ as it helps with the LAN segregation.  I can open up DMZ pinholes in my IPCop firewall. but so far this hasn't helped.

When a SAIL box is in server-only mode is it then unable to have remote phones register with it?

Thanks again

Norrie