[cisco-voip] TFTP Backup error

[Ed Leatherman]

Perfect! Thanks Florian

On Thu, Dec 29, 2011 at 2:58 PM, Florian Kroessbacher <
florian.kroessbacher at gmail.com> wrote:

> hy
>
> i think file list tftp * detail
>
> should do the job from cli
>
>
> Am 29.12.2011 um 20:47 schrieb Ed Leatherman <ealeatherman at gmail.com>:
>
> Anyone know if there is a way to see file sizes on files in TFTP file
> management or from CLI?
>
> On Thu, Dec 29, 2011 at 2:21 PM, Ed Leatherman <ealeatherman at gmail.com>wrote:
>
>> Worked with TAC for about 3 hrs on this today, nailed down the issue
>> though.
>>
>> So far I have tried with the lastest packaged build of openSSH for
>> ubuntu, as well as a built-from-scratch lastest stable build of openSSH
>> from openssh.org.
>>
>> Also, I've tried titanFTD and freeftpd on my local workstation and on a
>> windows server sitting on call manager publishers subnet - all attempts
>> both with openSSH and the windows packages fail at the exact same spot,
>> TFTP on my SUB-C server.
>>
>> Culprit: CSCsm21349 - DRS Direct Backup stuck for freeFTPd server when
>> tar size grow over 1GB
>>
>> Sure enough, in all cases the TFTP tarball failed at .99 GB.
>>
>> This bug report is somewhat misleading, as the workaround is to use
>> openSSH (which was what I started out with). None of the SFTP servers that
>> I tried worked, although I didn't try the cygwin flavor of openSSH. I also
>> have not tried the GlobalScape SFTP server which is supposedly Cisco
>> Certified (also it looks like its about $1200 or so with the SFTP module).
>> I suggested to the engineer that she should consider updating the bug
>> detail to indicate this issue could arise with other SFTP packages.
>>
>> Workaround for me was to just reduce the size of the TFTP directory on
>> the affected server. What triggered this issue in the first place was
>> installing Cius firmware packages willy-nilly, so i just removed some of
>> the older *cius* files using the cmplatform page on the particular servers.
>> I got the file size down to about 700MB and the backup worked fine.
>>
>> Now I get to figure out why TFTP is not even registered with DRF for my
>> primary TFTP server. Hasn't been giving an error since DRF isn't even
>> trying to back it up, I just noticed it was missing today.
>>
>>
>>
>> On Wed, Dec 14, 2011 at 2:45 PM, Ed Leatherman <ealeatherman at gmail.com>wrote:
>>
>>> Tried reinstalling that firmware that was showing up at the end of the
>>> log and restarting TFTP, then running a manual backup, still get the
>>> error. Looking in the backup directory on the SFTP server, it still
>>> looks like its writing all the files (even the tar for the TFTP server
>>> in question).
>>>
>>> I opened a TAC case on this one will follow up with what we find for
>>> posterity.
>>>
>>> On Wed, Dec 14, 2011 at 12:12 PM, Ryan Ratliff <rratliff at cisco.com>
>>> wrote:
>>> > In that case unless there's more detail in the drf backup.log I think
>>> a TAC
>>> > SR is the way to go.  I found a few old bugs with similar signatures
>>> but
>>> > they should be resolved in any version of CUCM that works with Cius.
>>> >
>>> > -Ryan
>>> >
>>> > On Dec 14, 2011, at 11:57 AM, Ed Leatherman wrote:
>>> >
>>> > on the tftp CM server, 35% used on active, less used on the other
>>> > partitions.
>>> > on the SFTP server, looks like i've got 60 gigs or so of free space, I
>>> > dont think its an issue there. it should be rotating out a backup
>>> > anyway.
>>> >
>>> > On Wed, Dec 14, 2011 at 10:15 AM, Ryan Ratliff <rratliff at cisco.com>
>>> wrote:
>>> >
>>> > What's your disk space look like both on the sftp server and CUCM?
>>> >
>>> >
>>> > -Ryan
>>> >
>>> >
>>> > On Dec 14, 2011, at 8:40 AM, Ed Leatherman wrote:
>>> >
>>> >
>>> > CM Backup fails with error in the TFTP backup log:
>>> >
>>> >
>>> > <big list of files here>
>>> >
>>> > ./pkg.cius.9-2-1SR1-19SEC.tgz
>>> >
>>> > tar: -: Wrote only 4096 of 10240 bytes
>>> >
>>> > tar: Error is not recoverable: exiting now
>>> >
>>> > Backup command [cd /usr/local/cm/tftp/; sudo nice -n 19 tar -cvf - .
>>> >
>>> > 2>> /common/drf/backup.log | sudo nice -n 19 gzip -c ] failed
>>> >
>>> > ----> BEGIN Standard Output
>>> >
>>> >
>>> > ----> END Standard Output
>>> >
>>> >
>>> >
>>> > ----> BEGIN Standard Error
>>> >
>>> >
>>> > gzip: stdout: Broken pipe
>>> >
>>> >
>>> > ----> END Standard Error
>>> >
>>> >
>>> > This is only one 1 server, TFTP on other nodes complete successfully
>>> >
>>> > (all other services complete successfully in fact)
>>> >
>>> >
>>> > Thoughts? looks like tar fails, not sure why though. TAC case material?
>>> >
>>> >
>>> > --
>>> >
>>> > Ed Leatherman
>>> >
>>> > _______________________________________________
>>> >
>>> > cisco-voip mailing list
>>> >
>>> > cisco-voip at puck.nether.net
>>> >
>>> > https://puck.nether.net/mailman/listinfo/cisco-voip
>>> >
>>> >
>>> >
>>> >
>>> >
>>> >
>>> > --
>>> > Ed Leatherman
>>> >
>>> >
>>>
>>>
>>>
>>> --
>>> Ed Leatherman
>>>
>>
>>
>>
>> --
>> Ed Leatherman
>>
>>
>
>
> --
> Ed Leatherman
>
> _______________________________________________
> cisco-voip mailing list
> cisco-voip at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>
>


-- 
Ed Leatherman
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/cisco-voip/attachments/20111229/5137548c/attachment.html>