[cisco-voip] CUCM Upgrade failure...

[Jonathan Charles]

Yep, the write-cache error cleared on the reboot... my guess is that was
what was killing it and the reboot fixed that...


On Wed, Nov 18, 2015 at 8:31 AM, Ryan Huff <ryanhuff at outlook.com> wrote:

> Hi Jonathan,
>
> Glad to hear a reboot on the publisher seems to have fixed your initial
> issue, I thought it might. You never know with reboots on those old MCS
> boxes though, so having a good DRS and recover plan prior is key.
>
> I suspect, and if you are so inclined, "show hardware" on the CLI of that
> pub would show that write-cache is enabled once again.
>
> Regarding the failure to connect to subscribers; one of the metrics used
> during UCOS installation to determine connectivity and reachability is
> response time and network visibility.  So if something takes to long to
> respond, the installer will say it isn't reachable but you can choose to
> ignore and installation will continue. Any latency deficit may, and likely
> will, cause issues in the cluster post install, but you can still complete
> the installation.
>
> Another thought would be visibility, if the installer can't see all the
> hops between nodes (assuming nodes are on different segments) it will say
> it can't connect during installation but you can choose to ignore and
> installation will continue. Routing may work just fine between the nodes
> but the installer will error if there are blind hops between nodes or it
> can't see the complete end to end path between nodes (a traceroute between
> nodes is a quick way to check). This is another item that may cause issues
> post install.
>
> -Ryan
>
> Sent from my T-Mobile 4G LTE Device
>
>
> -------- Original message --------
> From: Jonathan Charles
> Date:11/18/2015 9:02 AM (GMT-05:00)
> To: Dave Wolgast
> Cc: cisco-voip at puck.nether.net
> Subject: Re: [cisco-voip] CUCM Upgrade failure...
>
> A reboot of the Publisher fixed the time out issue.... after that the
> migration was successful... interestingly, the installation of subscribers
> fails due to 'no network connection to the node'.. however, you can open
> its webpage, and its media resources are registered... I just hit
> continue... dbreplication looks fine.
>
> Jonathan
>
> On Tue, Nov 17, 2015 at 2:30 PM, Dave Wolgast <dwolgas1 at rochester.rr.com>
> wrote:
>
>> On Tue, Nov 17, 2015 at 2:44 PM Heim, Dennis <Dennis.Heim at wwt.com> wrote:
>>
>>> Are you running the latest rev of PCD? I ran into the issue, and I
>>> believe upgrading to the latest rev resolved the issue.
>>>
>> Thanks Dennis. A colleague suggested that as well and I am in the process
>> of upgrading to PCD 11 now.
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/cisco-voip/attachments/20151118/91b7874b/attachment.html>