Ganglia builds are currently using v0.3.5 of CK

This version is troublesome in the i386 and ARM builds on Debian buildd
machine.  This impacts the availability of the CK dependency on both
Debian and Ubuntu.

Newer versions exist - has anybody tested a newer CK version with
Ganglia?  Is there any enthusiasm for using a newer version or any known
reason not to do so?

The travis build is hardcoded to ck-0.3.5 but I will shortly tweak it to
follow whatever version is available in Debian sid




On 06/04/14 02:42, Aaron M. Ucko wrote:
> Source: ck
> Version: 0.3.5-1
> Severity: important
> Justification: fails to build from source
> 
> The i386 build of ck failed the bytelock check:
> 
> https://buildd.debian.org/status/fetch.php?pkg=ck&arch=i386&ver=0.3.5-1&stamp=1396650656
> 
>   ----[ Testing bytelock....
>   make[3]: Entering directory 
> `/«PKGBUILDDIR»/regressions/ck_bytelock/validate'
>   ./validate 8 1
>   Creating threads (mutual exclusion)...done
>   Waiting for threads to finish correctness regression...ERROR [RD:110]: 8 != > 0
>   make[3]: *** [check] Error 1
>   make[3]: Leaving directory `/«PKGBUILDDIR»/regressions/ck_bytelock/validate'
> 
> FWIW, the kfreebsd-i386 build had no such problem; however, it ran the
> test with a different first argument (CPU core count?):
> 
> https://buildd.debian.org/status/fetch.php?pkg=ck&arch=kfreebsd-i386&ver=0.3.5-1&stamp=1396650149
> 
>   ----[ Testing bytelock....
>   make[3]: Entering directory 
> `/«PKGBUILDDIR»/regressions/ck_bytelock/validate'
>   ./validate 2 1
>   Creating threads (mutual exclusion)...done
>   Waiting for threads to finish correctness regression...done (passed)
>   make[3]: Leaving directory `/«PKGBUILDDIR»/regressions/ck_bytelock/validate'
> 
> Could you please take a look?
> 
> Thanks!
> 

------------------------------------------------------------------------------
_______________________________________________
Ganglia-developers mailing list
Ganglia-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/ganglia-developers

Reply via email to