Re: [bgl-discuss] machine locked

[Steven Pieper <pieper@xxxxxxxxxxx>]

I have been away for a week and am not sure of the final resolution
of the lengthy discussion that followed this but here are my comments:

1)  I sometimes want to produce lots of debugging output on stderr and 
8M is much too small.

2)  It is extremely useful to get stderr directly written in our home
file system (with names we specify) so that we can monitor jobs as
they run.  The queued returning of this output is not really nice
and adds the possibility of loosing it when something goes wrong
with the machine or scheduler.

steve


>>> I figured out and fixed what was happening. Your job produced 49M of
>>> output. We have a limit set at 8M, and the recovery case had a bug in
>>> it. I have removed the limit for now, but we should figure out what a
>>> reasonable one is. (we had a user on chiba generate 800M of output
>>> for a few jobs, and quickly ran into problems). This fix will kick in
>>> once i restart the queue manager.
>>>  -nld
>>> 
>>> - --------------------------------------------------------------------
>>> To add or remove yourself from this mailing list, use the 'notifyme'
>>> command on any BGL machine. To remove: notifyme -n, to add: notifyme -y.

- --------------------------------------------------------------------
To add or remove yourself from this mailing list, use the 'notifyme'
command on any BGL machine. To remove: notifyme -n, to add: notifyme -y.