This has been bothering me for quite some time. Resin is one of the best application servers I’ve come across. But to get the best performance out of it, you need to recompile it for your platform. While this is no problem on Linux it never quite worked out on Solaris for me when using SunStudio 12. But now I finally made it happen. Here’s how.
With the recent update of the Sun Studio 12 compiler set I started seeing errors like this when compiling for example MySQL from source:
ld.so.1: exe: fatal: relocation error: R_AMD64_32:
Turns out that it was Zlib 1.2.5 causing the problem.
Every once in a while I need to use dynamic variables in shell scripts for doing things like
but unfortunately it won’t work like shown in the example above.
What will work however is this:
Ah well. Compiling software on Solaris could be really easy if it weren’t for the small problems all the time. Today I was faced with
ld: fatal: file values-Xa.o: open failed: No such file or directory
when trying to compile Subversion 1.6.5 on OpenSolaris.
This is what I received today after offering a Solaris build environment to the guys from IonCube so they could build a version of their loader for my preferred web server OS:
If you like to monitor your machines using SNMP and find yourself on a machine running OpenSolaris, chances are that there is
already installed. But there’s no service or init script to actually run it automatically. In that case, just install the necessary package from the OpenSolaris repostiory:
pkg install SUNWsmmgr
Every time I connected to a Solaris machine from my Mac using ssh I ran into the dreaded “WARNING: terminal is not fully functional” problem. The reason is that the Mac sets its terminal type to “xterm-color” which isn’t known to Solaris. After fiddling with the termcap file with no success, I found the solution buried in the various tips on Phil’s site.
Just copy the xterm-color file provided by Phil (local copy) to the /usr/share/lib/terminfo/x/ directory. Done. Maybe you have to logout and login again to make it work, but that’s it.
Well, it’s over one and a half years that I first postedÂ some hintsÂ about tuning the performance of a Sun web server. Now it seems that I found what looks like the optimum settings for the machines I’m currently watching over:
That’s mostly consistent with what Jens S. VÃ¶ckler writes onÂ his site, just a bit more condensed.