This appears an architectural flaw in P-1 I guess. We see it
all the time -
whenever a 'container' hosting Primary CMA's goes down, and
the HA CMA's are
made active, upon Primary CMA's container coming back
online, both the CMAs
remain active - leading to 'collision' status and we have to
manually set it
right for 100's of CMAs.
Now coming to the global MDS level, both the Primary MDS
Manager and
Secondary MDS Managers have similar architectural design as
exists for the
CMAs at local level. I have seen the same behavior happen as
you are seeing
at the MDS level in our env. Couple of things need be
ensured:
1. What is the 'status' - is it in collision or
synchronized? You need to
ensure if any changes that were made on Sec Manager have
been sync'd to
Primary.
2. Once that is ensured, log into your Primary Manager MDS
global view,
right click on the Secondary Manager and change to
'standby'.
We had raised this question w/ CP earlier and there is no
solution to the
problem when the Primaries' elements go down. One has to
manually fix them
unless there is something in NGX (some version), they have
changed the
design.
Hope this helps.
Rajeev
On 7/15/07, cisco4ng <cisco4ng yahoo.com> wrote:
>
> I have a P-1 question for gurus in the forum.
>
> I have the following:
>
> P-1 NGx R61 with hfa_01 PRIMARY MANAGER on SPLAT.
> The ip address of this box is 192.168.1.9/24.
> Hostname of this box is LAB-Mgr
>
> P-1 NGx R61 with hfa_01 SECONDARY MANAGER on SPLAT.
> The ip address of this box is 192.168.1.7/24.
> Hostname of this box is LAB-Mgr-Sec
>
> P-1 NGx R61 with hfa_01 CONTAINER on SPLAT.
> The ip address of this box is 192.168.1.10/24.
> Hostname of this box is LAB-Cntr
>
> Everything is running EVAL license.
>
> User(s) can log into P-1 via MDG. Everyone
> is logging into the MDG via the Primary Manager.
> Under the High Availability, it shows that
> the LAB-Mgr is Active and the LAB-Mgr-Sec is
> "standby".
>
> For testing purposes, I reboot the LAB-Mgr box
> Now everyone logs into the P-1 via the LAB-Mgr-Sec
> and under the "High Availability", it shows
> the LAB-Mgr-Sec as "Active".
>
> When the LAB-Mgr comes back online, it also
> shows as "Active" as well, and it remains
> that way for the past 3 days now. In other words,
> I have both LAB-Mgr and LAB-Mgr-Sec are both
> showing as "active". By the way, I only
have
> 2 CMAs on the container LAB-Cntr box.
>
> Does it mean I have a problem because both
> Manager boxes show up as "active/active"
instead
> of "active/standby"?
>
> All of these boxes are connected to the same switch
> and they are also on the same network so latency
> can not be an issue, right?
>
> Can someone with P-1 experience explain this?
>
> Thanks.
>
>
>
>
>
>
> ---------------------------------
> Pinpoint customers who are looking for what you sell.
>
> ---------------------------------
> Take the Internet to Go: Yahoo!Go puts the Internet in
your pocket: mail,
> news, photos & more.
>
> =================================================
> To set vacation, Out-Of-Office, or away messages,
> send an email to LISTSERVamadeus.us.checkpoint.com
> in the BODY of the email add:
> set fw-1-mailinglist nomail
> =================================================
> To unsubscribe from this mailing list,
> please see the instructions at
> http:
//www.checkpoint.com/services/mailing.html
> =================================================
> If you have any questions on how to change your
> subscription options, email
> fw-1-ownerts.checkpoint.com
> =================================================
>
=================================================
To set vacation, Out-Of-Office, or away messages,
send an email to LISTSERVamadeus.us.checkpoint.com
in the BODY of the email add:
set fw-1-mailinglist nomail
=================================================
To unsubscribe from this mailing list,
please see the instructions at
http:
//www.checkpoint.com/services/mailing.html
=================================================
If you have any questions on how to change your
subscription options, email
fw-1-ownerts.checkpoint.com
=================================================
|
