Re: Odd error in testing

2020-01-23 Thread Andrzej Białecki
The metrics history API can generate graphs and then indeed it uses the built-in support in RRD4j for this, which in turn uses AWT. The DejaVuSansMono.ttf font is packaged inside rrd4j-*.jar so it doesn’t have to exist in the filesystem. > On 23 Jan 2020, at 11:38, Jan Høydahl wrote: > >

Re: Odd error in testing

2020-01-23 Thread Jan Høydahl
Appears that rrd4j always tries to look for a font in the static initialization of RrdGraphConstants: https://github.com/rrd4j/rrd4j/blob/master/src/main/java/org/rrd4j/graph/RrdGraphConstants.java#L297

Re: Odd error in testing

2020-01-23 Thread Shalin Shekhar Mangar
I saw this as well on a Ubuntu Linux box running openjdk 11.0.6. The failures were 100% reproducible. I followed https://github.com/AdoptOpenJDK/openjdk-build/issues/682 and see that a suggestion is to install fontconfig package. Unfortunately this box is running an old version and I must upgrade

Re: Odd error in testing

2018-12-11 Thread David Smiley
I wonder why RRD4J is doing any AWT stuff at all. I thought we were just using it to *hold* data, not to visualize. ab? On Sun, Dec 9, 2018 at 3:27 AM Erick Erickson wrote: > I reformatted the entire disk on my old machine and installed Mojave > fresh then ran tests. > >

Odd error in testing

2018-12-08 Thread Erick Erickson
I reformatted the entire disk on my old machine and installed Mojave fresh then ran tests. MetricsHistoryIntegrationTest.testGet fails on my old (reformatted, freshly installed Mojave) with the bits below. It does _not_ fail on any other machine. Weird bits about: Caused by: