cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From David Crossley <cross...@apache.org>
Subject Re: ForrestBot build for cocoon-docs FAILED
Date Thu, 23 Mar 2006 03:18:41 GMT
Ross Gardler wrote:
> Bruno Dumon wrote:
> >David Crossley wrote:
> >>David Crossley wrote:
> >>>Bruno Dumon wrote:
> >>>
> >>>>BTW, this time not because the zone was restarted. Not sure what the
> >>>>problem is though.
> >>>
> >>>I investigated a bit yesterday, but cannot see what
> >>>is the problem. Notice that today there are less breaks
> >>>than yesterday. However, looking at the cocoon-docs mailing
> >>>list diffs from Daisy i cannot see any relevant changes.
> >>>
> >>>http://forrest.zones.apache.org/ft/build/cocoon-docs/broken-links.xml
> >>>
> >>>It complains about Daisy IDs 786 and 655 which are
> >>>navigation documents. Surely Forrest's Cocoon had
> >>>already processed those resources.
> >>>
> >>>Strange. I am stumped. Ross seems busy with other stuff.
> >>
> >>Curiouser and curiouser. At the next 12-hourly run
> >>there is only one breakage and for a different ID.

Then the next day the issue was worse.

> >FYI: I've had a look at the Daisy log files, and don't see anything
> >there. The server has also been continuously up.
>
> I've not looked into this in any detail. I'm swamped right now. But it 
> looks to me like there are intermittent network problems between the two 
> zones. I've noticed this failure come and go without anyone touching our 
> Forrestbot or the sources.
> 
> Is this possible, as far as I am aware they are on the same physical 
> machine, but separate virtual machines.

Same machine yes. Using the "Zones" of the Solaris 10 OS.

> Perhaps there is a time out occuring. Is there a way of increasing the 
> timeout on files being generated from http: sources? Not a solution, but 
> it would be a workaround.

I don't know if it proves anything, but just completed
a local build and it worked. This is horrendously slow
for some reason (takes 25 seconds per page).

-David

Mime
View raw message