![]() | |||||||||||||
Messages in this thread |
On Mar 14, 2006, at 12:16:23, Gene Heskett wrote: > On Tuesday 14 March 2006 10:05, Phillip Susi wrote: >> Or look in your /var/log/kern.log file instead of asking dmesg. >> dmesg just dumps the kernel ring buffer which is of finite size. >> The entire contents should be logged to /var/log/kern.log. > > You've got to have /var checked and mounted to be able to do that > log write, if the buffer overflows before that, then the head end > of the dmesg dump to the /var/log/dmesg file is lost forever. > > There is a line that can be changed, in xconfig or by hand, to > control the memory allocated for this ring buffer. > > Finally found it in the xconfig display, left panel line=kernel > hacking, right panel its under kernel debugging and shows only if > thats checked, double click on the line that says : kernel log > buffer size and enter a one digit increment from whats there now, > maybe 2, I have mine set for 16. Your default may be as low as 14, > why I have NDI because 16k sure as heck isn't enough if something > gets chatty. To continue this point; on my desktop I have root/var/tmp/vicepa-on- LVM-on-RAID5, boot-on-RAID1, and swap-on-RAID1, so I would easily overflow the default SMP dmesg buffer size in messages well before syslogd/bootlogd got started. I finally ended up having to increment the default by 3 in order to have the boot messages still available after booting. It would be nice if we could quiet down some of the more excessively verbose kernel messages, there's a lot of mostly- irrelevant spew that chews up log buffer space. Cheers, Kyle Moffett - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ | ||||||||||||
| Last update: 2006-03-14 18:31 [from the cache] ©2003-2008 | |||||||||||||