'Vertical size too small' error from rt-email-dashboards

Hello!

We’re getting an error from our daily rt-email-dashboards Cron job:

 [Fri Nov 21 06:00:17 2014] [error]: Vertical size too small at 

/opt/rt4/share/html/Search/Chart line 185.

 Stack:
   [/opt/rt4/share/html/Search/Chart:185] 

(/opt/rt4/sbin/…/lib/RT/Interface/Web/Handler.pm:208)
[Fri Nov 21 06:00:17 2014] [error]: Caught exception: An internal
RT error has occurred. Your administrator can find more details in RT’s
log files. (/opt/rt4/sbin/…/lib/RT/Dashboard/Mailer.pm:118)

As far as I know, nothing’s been changed in our RT configuration since
when it started happening - though perhaps a user has made changes to
one of their subscribed dashboards so that it’s now too big?

Someone else ran into the same issue at
http://www.gossamer-threads.com/lists/rt/users/107020, but there was no
resolution. Googling for just “Vertical size too small”, it looks like
this is an error from Perl’s GDGraph library, with a suggestion to
adjust the correct_width attribute on the plot_graph call:
http://search.cpan.org/dist/GDGraph/Graph.pm But in order to test
whether this fixes it, it would be nice to have a repeatable test case.

Running rt-emails-dashboads --dryrun --all I see there are about 20
dashboard subscriptions set up. How can we figure out which one(s) are
giving the error, so that we can come up with a reliable test case (and
without spamming users with emails :slight_smile: )?

Cheers,
Matthew

Matthew Rahtz
Informatics Systems Group
Wellcome Trust Sanger Insitute

The Wellcome Trust Sanger Institute is operated by Genome Research
Limited, a charity registered in England with number 1021457 and a
company registered in England with number 2742969, whose registered
office is 215 Euston Road, London, NW1 2BE.

We’re getting an error from our daily rt-email-dashboards Cron job:

[Fri Nov 21 06:00:17 2014] [error]: Vertical size too small at

/opt/rt4/share/html/Search/Chart line 185.

I’m surprised you get the error in cron but not

Running rt-emails-dashboads --dryrun --all I see there are about 20
dashboard subscriptions set up. How can we figure out which one(s) are
giving the error, so that we can come up with a reliable test case (and
without spamming users with emails :slight_smile: )?

I would expect you to get the error during the chart render unless
it’s very sensitive to timing.

You can set $LogToScreen ($LogToSTDERR on 4.2) to debug so that your
cron job will spew more information about which dashboard is being
processed when the error is run. After that, rt-attributes-viewer on
the dashboard and relevant chart will be interesting.

-kevin