[MITgcm-devel] Re: today testing

[Ed Hill]

On Thu, 12 Oct 2006 10:42:16 -0400 (EDT) marshall-admin at techsquare.com
wrote:

> hello ed-
> 
> can you send mail next time ? 
> turns out that all the nfs mounts
> at forge were gone, which would 
> explain this. 
> 
> this will become even more important
> the longer you are absent from mit,
> as things do change (however slowly).
> 
> thanks, and can you remove the substitute
> munpack now too ? 


Hi Greg, Chris, etc.

I put the "substitute" munpack in:

  ~edhill/local/bin/munpack

and then modified the script (called hourly from edhill's cron on
forge) that does the unpacking so that it uses that one instead of
/usr/local/bin/munpack.  Thats all.  Its *not* an actual system
install.

And, yes, I *totally* agree that things will change and that the test
bits need to have someone looking over them.  They can run smoothly
with no help for weeks or they can die suddenly when an NFS mount
fails, etc.  They just need a little love occasionally.

So the real fix here is for someone new to take over ownership of the
testing scripts.  Why don't you folks (esp. Chris) find that "lucky
someone" and then have them contact me and I'll help him/her make the
process as painless as possible.  And, honestly, it won't be too hard.
The scripts are currently tied to my account and email address on
forge ("edhill at mitgcm.org") but it'll easy enough to make some small
edits so that they'll use a new account (and associated email address)
that could even be a completely neutral "role" account such as
"testing at mitgcm.org" or similar.

Ed

-- 
Edward H. Hill III, PhD  |  ed at eh3.com  |  http://eh3.com/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
URL: <http://mitgcm.org/pipermail/mitgcm-devel/attachments/20061012/788806d2/attachment.sig>