[Box-Admins] Re: box4 is inaccessible

[Chris Muller]

It is running a much newer image (4.5 instead of 3.11), I don't know if it
can support the special chars or not, but as I went to test it, I found teh
cause of the box4 crash, and a number of other problems:

   - my own "repcopy" script which is running on an hourly cron is getting
some error, but leaving the process (squeak iamge) running.  My bet is box4
ran out of memory and crashed.  I just killed about 50 of those processes.
The script redirects stdout and stderr to "repcopy.log" and "repcopy.err",
respectively.  However, upon the next scheduled run, I found the process
hung again but those two files were empty.  However, there was a
SqueakDebug.log file...

  - The SqueakDebug.log file shows *Squeak-Version-kfr.4712* is triggering
the Warning, "About to serialize empty mcz".  This version has a comment
of, "Messed up last version, try again", and so is a piece of litter in our
ancestry and trunk.  I again ask y'all to support me on this littering
issue and help me address it with community.  Please!

  - The weird thing is, the script is wrapped in a Smalltalk>>#run:, which
handles all Warnings by resuming them.  So, I don't know why it is halting..

As for production with the special-char:

   - the object model for production source.squeak.org is stored in file

            /srv/box2/home/squeaksource/ss/data.obj

    but it is currently timestamped April 20th, 2015.  So any changes to
that model since that time, including the addition of Stephan and Patrick
are probably only in the production image
(/srv/box2/home/squeaksource/Squeak3.11-8824-SS.image).

   - I have just downloaded that image
/srv/box2/home/squeaksource/Squeak3.11-8824-SS.image and exported a new
"data.obj" file.

Next I will make a new method in ss-deploy.image to refresh the Magma db
with the model from data.obj without having to reload all the versions.  I
will then know whether it can handle the special characters in people's
names..



On Tue, Dec 1, 2015 at 2:35 PM, Levente Uzonyi <leves at caesar.elte.hu> wrote:

> You were the one who set up a newer image with Magma backend on box4,
> though that image hasn't synchronized the list of users, so it's web
> interface still works. I assume, it would break, too.
>
> Levente
>
>
> On Tue, 1 Dec 2015, Chris Muller wrote:
>
> Would a newer image and VM solve this?
>>
>> Why do we continue to run on this old image and VM?
>>
>>
>>
>> On Tue, Dec 1, 2015 at 7:56 AM, David T. Lewis <lewis at mail.msen.com>
>> wrote:
>>       On Tue, Dec 01, 2015 at 12:42:33PM +0100, Levente Uzonyi wrote:
>>       > That's probably because someone has registered with a name
>> containing wide
>>       > characters, so the whole page is served as a WideString. This bug
>> has
>>       > happened a few times before, and I don't remember what the
>>       > temporary solution was, but I think Dave knows.
>>       >
>>       > Levente
>>
>>       That is probably it. Note that St??phane Rollandin (spfa) has just
>> registered
>>       a new account. That is probably showing up on the source.squeak.org
>> home
>>       page, and blowing up when the wide string is displayed.
>>
>>       I see the same thing on squeaksource.com from time to time. If we
>> do
>>       nothing, it will probably "fix" itself in a little while (after
>> Stef's
>>       activity becomes old news and scrolls off the home page).
>>
>>       If the problem does not go away by itself, may need to fix it in the
>>       source.squeak.org image, which involves finding the offending
>> WideString
>>       and patching it to be displayable.
>>
>>       The good news is that we are adding a more diverse range of
>> developers
>>       to source.squeak.org :-)
>>
>>       Dave
>>
>>
>>
>>
>>       >
>>       > On Tue, 1 Dec 2015, Bert Freudenberg wrote:
>>       >
>>       > >source.squeak.org looks weird.
>>       > >
>>       > >- Bert -
>>       > >
>>       > >
>>       > >
>>       > >>On 01.12.2015, at 00:12, Levente Uzonyi <leves at elte.hu> wrote:
>>       > >>
>>       > >>Thanks whoever requested the restart. The machine seems to have
>> four
>>       > >>times as much RAM as before. I've hopefully restarted all
>> services.
>>       > >>Should anything be missing, let me know!
>>       > >>
>>       > >>Levente
>>       > >>
>>       > >>On Mon, 30 Nov 2015, Levente Uzonyi wrote:
>>       > >>
>>       > >>>Hi All,
>>       > >>>
>>       > >>>box4 seems to have been down since yesterday. Box3 is fine, so
>> hopefully
>>       > >>>a restart will bring box4 back up.
>>       > >>>The problem is that only the SFC has access to the box, and
>> only they
>>       > >>>can request a restart.
>>       > >>>So please ask them to do so.
>>       > >>>
>>       > >>>Thanks,
>>       > >>>Levente
>>       > >>>
>>       > >
>>       > >
>>
>>
>>
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.squeakfoundation.org/pipermail/box-admins/attachments/20151201/55b11c12/attachment-0001.htm