Date: Tue, 31 May 2011 15:43:22 +0200 From: Matthias Andree <matthias.andree@gmx.de> To: freebsd-ports@freebsd.org Subject: Re: ports/153337: print/acroread9: terminate called after throwing an instance of 'RSException' Message-ID: <4DE4F07A.100@gmx.de> In-Reply-To: <19940.54861.206263.982241@jerusalem.litteratus.org> References: <201105291730.p4THUkkA049372@freefall.freebsd.org> <4DE28663.5080408@mail.zedat.fu-berlin.de> <20110530.030010.439410720786741904.hrs@allbsd.org> <4DE34479.4080605@mail.zedat.fu-berlin.de> <4DE3467E.3090000@gwdg.de> <4DE36B61.6080901@zedat.fu-berlin.de> <19939.34490.933805.678124@jerusalem.litteratus.org> <4DE38A0A.5060504@gmx.de> <19939.38697.202103.902396@jerusalem.litteratus.org> <4DE397D7.5000707@gmx.de> <19940.54861.206263.982241@jerusalem.litteratus.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Am 31.05.2011 13:51, schrieb Robert Huff: > > Matthias Andree writes: > >> > May 30 07:56:37 jerusalem kernel: KLD linux_adobe.ko: depends on kernel - not available or version mismatch >> > May 30 07:56:37 jerusalem kernel: linker_load_file: Unsupported file type >> > May 30 07:57:48 jerusalem huff: /usr/local/etc/rc.d/linux_adobe: WARNING: Unable to load kernel module /usr/local/libexec/linux_adobe/linux_adobe.ko >> >> "version mismatch" looks likely here - are you sure that your >> running kernel and the kernel sources are in synch? > > Quite sure they're not. Kernel sources get updated every day > at (local) midnight; they're used to upgrade the kernel ... usually > every month (sometimes longer), but sometimes quickly if a problem > if discovered that update corrects. So that's the cause. Simple solution: rebuild your kernel source from /usr/src/sys, reboot, and rebuild and reinstall all modules with the same kernel sources in place. I believe something like this isn't too hard to grasp from the "staying current" handbook sections. If it's not, feel free to propose a wording. Don't update or build new kernel modules until you're ready to upgrade your kernel and reboot. >> If not, that's where the mismatch comes from. > > This is the second program - the other is sysutils/lsof - that > makes this (to my mind unwarranted) assumption. > Recognizing this unlikely to change, can this expectation be > documented so it is announced at compile time? The assumption is neither unwarranted nor surprising. 1. You are compiling kernel modules, i. e. stuff that meddles with /internal/ kernel interfaces (else it were a regular application, not a kernel module) 2. You are also using -CURRENT, which is a moving target, by definition, and as documented. (The same applies to -STABLE BTW. - if you want really stable code, stick to the -RELEASE-s.) You need to make sure that the kernel sources (including headers) of your currently running kernel and the modules to be loaded are compatible, because the header files in the kernel sources are what defines the interfaces.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4DE4F07A.100>