Re: [leaf-devel] Bering 1.2: linuxrc, root.dev.mk and logging
Jon Clausen wrote: Am I reading /var/lib/lrpkg/root.dev.mk correctly? AFAICT everything it does sends any output to /dev/null (except possibly the cd-rom stuff right at the end). Assuming that it is meant to work silently (ie no logging), then there shouldn't be a problem with swapping the code in /linuxrc so that /var/log gets mounted *after* devices are created? I ask because: Wanting to log to harddisk, I had to do the above. It worked nice (thread on leaf-user ends with: Harddisk: device deceased) apart form the fact that the disk died. I subsequently got a personal request to explain how I did it. This all made me think that it would be nice to 'formalize' the functionality. So I concocted some code to parse a log_dev=/dev/hdXn:fstype -statement in /proc/cmdline. Obviously an actual mount can't take place before the device to mount is created. The code seems to do what it's supposed to, but before I submit it here for review, it seemed like a good idea to get it straight whether or not there would be a problem with mounting /var/log after root.dev.mk runs. There should be no problem making the root devices prior to mounting /var/log. In fact, in the latest init scripts (see Bering uClibc or my not-quite-released Bering-CD), the creation of block devices occurs prior to mounting the *ROOT* ramdisk (to allow reading of leaf.cfg and support run-time configuration of ramdisk size). -- Charles Steinkuehler [EMAIL PROTECTED] --- SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33 Save 50% off Retail on Ink Toner - Free Shipping and Free Gift. http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285 ___ leaf-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-devel
Re: [leaf-devel] Bering 1.2: linuxrc, root.dev.mk and logging
On Wed, 11 Aug, 2004 at 06:30:19 -0500, Charles Steinkuehler wrote: Jon Clausen wrote: snip The code seems to do what it's supposed to, but before I submit it here for review, it seemed like a good idea to get it straight whether or not there would be a problem with mounting /var/log after root.dev.mk runs. There should be no problem making the root devices prior to mounting /var/log. O.k. Thanks for confirming. In fact, in the latest init scripts (see Bering uClibc or my not-quite-released Bering-CD), the creation of block devices occurs prior to mounting the *ROOT* ramdisk (to allow reading of leaf.cfg and support run-time configuration of ramdisk size). Right. So if I contribute something to mount /var/log on disk, that would also work out-of-the-box-ish with those... (I guess?) Next on *my* agenda is to decide whether to make 1: A 'unified' log= that wil parse both - log=xM (to replace the original log_size=xM option) - log=/dev/hdXn:fstype or 2: Keep log_size, add log_dev and make them mutually exclusive (perhaps with a preference for log_size if both should be present in cmdline) I like 1: better, but maybe it's a bit too ambiguous? 2: on the other hand is probably simpler to write. Decisions, decisions... :) Comments very welcome /Jon -- Just say know! --- SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33 Save 50% off Retail on Ink Toner - Free Shipping and Free Gift. http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285 ___ leaf-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/leaf-devel