[rancid] Backing Up Cisco ACE Modules

Discussion:

Brown IV, Bill (MFT)

2009-11-30 18:41:50 UTC

Hello all,

I've been trying to backup the Cisco ACE but Rancid seems to time out during the connection. I think it has to do with the pager. On the ACE, every time I login the pager gets set back to its default. Logging in and then setting 'term length 0' only works for the session. After Rancid logs out, the pager gets reset.

Scenario: I can login to the device using clogin. Issuing a 'show run' from there reveals that the pager is set, (noted by the --more-- prompt). I set 'term length 0' and 'sho run' completes with no pager. Logging out and then back in reveals that the pager is once again set. Using the rancid script with -d, the last output is:

HIT COMMAND:<device>/Admin#show running-config
In WriteTerm: <device>/Admin#show running-config
HIT COMMAND: <device>/Admin#write term
In WriteTerm: <device>/Admin#write term
<device>: End of run not found
<device>: End of run not found

My assumption is that somehow Rancid is expecting the enable prompt but not getting it because of the pager. I looked at the rancid script but I'm just not proficient enough to know where to hard code a 'term length 0' and I definitely don't want to change the behavior for other devices, (I don't want to "mess up"). Does anyone have a piece of code that will accomplish this?

If possible, if you have it, I would really like to backup all the contexts within the ACE as well.

Thanks,
Bill

N***@guesswho.com

2009-11-30 18:56:00 UTC

Permalink

Bill,
I have been backing up our ACE modules for a while now. I had to seriously modify a script to make it work (I am not a developer so please excuse) and I am only concerned about the config and the fault tolerance info (this helps to know if one has failed over)

I will send the script off list but remember you need to add the appropriate lines to rancid-fe for the new device type:

elsif ($vendor =~ /^ace$/i) { exec('acerancid', $router); }

and I used "add autoenable" in .cloginrc.

Let me know if you have any questions.
Nick

From: rancid-discuss-***@shrubbery.net [mailto:rancid-discuss-***@shrubbery.net] On Behalf Of Brown IV, Bill (MFT)
Sent: Monday, November 30, 2009 1:42 PM
To: rancid-***@shrubbery.net
Subject: [rancid] Backing Up Cisco ACE Modules

Hello all,

I've been trying to backup the Cisco ACE but Rancid seems to time out during the connection. I think it has to do with the pager. On the ACE, every time I login the pager gets set back to its default. Logging in and then setting 'term length 0' only works for the session. After Rancid logs out, the pager gets reset.

Scenario: I can login to the device using clogin. Issuing a 'show run' from there reveals that the pager is set, (noted by the --more-- prompt). I set 'term length 0' and 'sho run' completes with no pager. Logging out and then back in reveals that the pager is once again set. Using the rancid script with -d, the last output is:

HIT COMMAND:<device>/Admin#show running-config
In WriteTerm: <device>/Admin#show running-config
HIT COMMAND: <device>/Admin#write term
In WriteTerm: <device>/Admin#write term
<device>: End of run not found
<device>: End of run not found

My assumption is that somehow Rancid is expecting the enable prompt but not getting it because of the pager. I looked at the rancid script but I'm just not proficient enough to know where to hard code a 'term length 0' and I definitely don't want to change the behavior for other devices, (I don't want to "mess up"). Does anyone have a piece of code that will accomplish this?

If possible, if you have it, I would really like to backup all the contexts within the ACE as well.

Thanks,
Bill

Michael Stefaniuc

2009-12-01 09:38:40 UTC

Permalink

Hi Bill,

I’ve been trying to backup the Cisco ACE but Rancid seems to time out
during the connection. I think it has to do with the pager. On the
ACE, every time I login the pager gets set back to its default. Logging
in and then setting ‘term length 0’ only works for the session. After
Rancid logs out, the pager gets reset.

you can try to use my patch from
http://www.shrubbery.net/pipermail/rancid-discuss/2008-October/003338.html

Scenario: I can login to the device using clogin. Issuing a ‘show run’
from there reveals that the pager is set, (noted by the --more--
prompt). I set ‘term length 0’ and ‘sho run’ completes with no pager.
Logging out and then back in reveals that the pager is once again set.
HIT COMMAND:<device>/Admin#show running-config
In WriteTerm: <device>/Admin#show running-config
HIT COMMAND: <device>/Admin#write term
In WriteTerm: <device>/Admin#write term
<device>: End of run not found
<device>: End of run not found
My assumption is that somehow Rancid is expecting the enable prompt but
not getting it because of the pager. I looked at the rancid script but
I’m just not proficient enough to know where to hard code a ‘term length
0’ and I definitely don’t want to change the behavior for other devices,
(I don’t want to “mess up”). Does anyone have a piece of code that will
accomplish this?

It's not the pager but rancid's detection of the end of the config file
which goes wrong.

If possible, if you have it, I would really like to backup all the
contexts within the ACE as well.

For now you have to give each context an IP and use that. I plan to add
that feature but I'm pretty busy lately.

bye
michael

--
Michael Stefaniuc Tel.: +49-711-96437-199
Consulting Communications Engineer Fax.: +49-711-96437-111
--------------------------------------------------------------------
Reg. Adresse: Red Hat GmbH, Otto-Hahn-Strasse 20, 85609 Dornach
Handelsregister: Amtsgericht Muenchen HRB 153243
Geschäftsführer: Brendan Lane, Charlie Peters, Michael Cunningham,
Charles Cachera

Remy Heiden

2009-12-01 12:35:59 UTC

Permalink

Strange..

We backup our ACEs with no trouble at all..
Just using the standard rancid script and separate management IP's on the
contexts.

No issues with pagers or end of config files...

I'll have a look to see if i can find some details..

Remy

Post by Michael Stefaniuc
Hi Bill,

Ive been trying to backup the Cisco ACE but Rancid seems to time out
during the connection. I think it has to do with the pager. On the
ACE, every time I login the pager gets set back to its default. Logging
in and then setting term length 0 only works for the session. After
Rancid logs out, the pager gets reset.

you can try to use my patch from
http://www.shrubbery.net/pipermail/rancid-discuss/2008-October/003338.html

Scenario: I can login to the device using clogin. Issuing a show run
from there reveals that the pager is set, (noted by the --more--
prompt). I set term length 0 and sho run completes with no pager.
Logging out and then back in reveals that the pager is once again set.
HIT COMMAND:<device>/Admin#show running-config
In WriteTerm: <device>/Admin#show running-config
HIT COMMAND: <device>/Admin#write term
In WriteTerm: <device>/Admin#write term
<device>: End of run not found
<device>: End of run not found
My assumption is that somehow Rancid is expecting the enable prompt but
not getting it because of the pager. I looked at the rancid script but
Im just not proficient enough to know where to hard code a term length
0 and I definitely dont want to change the behavior for other devices,
(I dont want to mess up). Does anyone have a piece of code that will
accomplish this?

It's not the pager but rancid's detection of the end of the config file
which goes wrong.

If possible, if you have it, I would really like to backup all the
contexts within the ACE as well.

For now you have to give each context an IP and use that. I plan to add
that feature but I'm pretty busy lately.
bye
michael
--
Michael Stefaniuc Tel.: +49-711-96437-199
Consulting Communications Engineer Fax.: +49-711-96437-111
--------------------------------------------------------------------
Reg. Adresse: Red Hat GmbH, Otto-Hahn-Strasse 20, 85609 Dornach
Handelsregister: Amtsgericht Muenchen HRB 153243
Geschäftsführer: Brendan Lane, Charlie Peters, Michael Cunningham,
Charles Cachera
_______________________________________________
Rancid-discuss mailing list
http://www.shrubbery.net/mailman/listinfo.cgi/rancid-discuss

--
==========================================================

"The smell, you know that gasoline smell, the whole hill. Smelled like...
victory."

(Lieutenant Colonel Bill Kilgore)

==========================================================
--

Remy Heiden

2009-12-01 12:43:19 UTC

Permalink

Actually, we do have a slight issue with thge ACEs...

Rancid ONLY collects the configs and not any of the other commands:

!RANCID-CONTENT-TYPE: cisco
!
!
!
!
!
!
!
!
!
!
^
% invalid command detected at '^' marker.

Generating configuration....
<SNIP>

As for now we only are interested in the configs this is no issue for us.

R.

Post by Remy Heiden
Strange..
We backup our ACEs with no trouble at all..
Just using the standard rancid script and separate management IP's on the
contexts.
No issues with pagers or end of config files...
I'll have a look to see if i can find some details..
Remy

Post by Michael Stefaniuc
Hi Bill,

you can try to use my patch from
http://www.shrubbery.net/pipermail/rancid-discuss/2008-October/003338.html

It's not the pager but rancid's detection of the end of the config file
which goes wrong.

If possible, if you have it, I would really like to backup all the
contexts within the ACE as well.

--
==========================================================
"The smell, you know that gasoline smell, the whole hill. Smelled like...
victory."
(Lieutenant Colonel Bill Kilgore)
==========================================================
--

Michael Stefaniuc

2009-12-01 13:08:06 UTC

Permalink

Post by Remy Heiden
Actually, we do have a slight issue with thge ACEs...

The "right" commands for the ACE aren't in and the IOS ones aren't known.

Post by Remy Heiden
!RANCID-CONTENT-TYPE: cisco
!
!
!
!
!
!
!
!
!
!
^
% invalid command detected at '^' marker.

This is actually a different bug for which I have just sent a patch:
http://www.shrubbery.net/pipermail/rancid-discuss/2009-December/004447.html

bye
michael

Post by Remy Heiden
Generating configuration....
<SNIP>
As for now we only are interested in the configs this is no issue for us.
R.
Strange..
We backup our ACEs with no trouble at all..
Just using the standard rancid script and separate management IP's
on the contexts.
No issues with pagers or end of config files...
I'll have a look to see if i can find some details..
Remy
On Tue, Dec 1, 2009 at 10:38 AM, Michael Stefaniuc
Hi Bill,

I’ve been trying to backup the Cisco ACE but Rancid seems to

time out

during the connection. I think it has to do with the pager.

On the

ACE, every time I login the pager gets set back to its

default. Logging

in and then setting ‘term length 0’ only works for the

session. After

Rancid logs out, the pager gets reset.

you can try to use my patch from
http://www.shrubbery.net/pipermail/rancid-discuss/2008-October/003338.html

Scenario: I can login to the device using clogin. Issuing a

‘show run’

from there reveals that the pager is set, (noted by the --more--
prompt). I set ‘term length 0’ and ‘sho run’ completes with

no pager.

Logging out and then back in reveals that the pager is once

again set.

HIT COMMAND:<device>/Admin#show running-config
In WriteTerm: <device>/Admin#show running-config
HIT COMMAND: <device>/Admin#write term
In WriteTerm: <device>/Admin#write term
<device>: End of run not found
<device>: End of run not found
My assumption is that somehow Rancid is expecting the enable

prompt but

not getting it because of the pager. I looked at the rancid

script but

I’m just not proficient enough to know where to hard code a

‘term length

0’ and I definitely don’t want to change the behavior for

other devices,

(I don’t want to “mess up”). Does anyone have a piece of code

that will

accomplish this?

It's not the pager but rancid's detection of the end of the config file
which goes wrong.

If possible, if you have it, I would really like to backup all the
contexts within the ACE as well.

For now you have to give each context an IP and use that. I plan to add
that feature but I'm pretty busy lately.