Hi: I ask cos I searched the archive and found nothing about cruft (http://www.mail-archive.com/cdwrite@other.debian.org/index.html). Searching in other places gives me a lot of links (Linux kernel files that I already have, or similar) but no definitive documentation / explanation.
I have been using some options in mkisofs to group CDs (-volset "string" -volset-size N -volset-seqno n). I read the man page and thought they where a good idea to store the relation among them. I also read the following page (to check other docs, just in case): http://www.y-adagio.com/public/standards/iso_cdromr/tocont.htm Now the fun, when I mount a CD with seqno above 1, I get in the logs: kernel: Warning: defective CD-ROM (volume sequence number 3). Enabling "cruft" mount option. I am currently using RH71, mkisofs is 1.13 (i686-pc-linux-gnu) and kernel is 2.4.2-2. I scanned the Linux src I have and found fs/isofs/inode.c around line 1251: /* * Disable checking if we see any volume number other than 0 or 1. * We could use the cruft option, but that has multiple purposes, one * of which is limiting the file size to 16Mb. Thus we silently allow * volume numbers of 0 to go through without complaining. */ if (inode->i_sb->u.isofs_sb.s_cruft == 'n' && (volume_seq_no != 0) && (volume_seq_no != 1)) { printk(KERN_WARNING "Warning: defective CD-ROM " "(volume sequence number %d). " "Enabling \"cruft\" mount option.\n", volume_seq_no); inode->i_sb->u.isofs_sb.s_cruft = 'y'; I think the Linux file is wrong (but I am not at expert, not even a kernel newbie). Also other things look rare, like testing that filesystem size is less than 1Gig, and as fix-all use the cruft option too (a radical one specially when I see no relation with the supposed problems, at least in the case of volset-seqno). I think the doc and the man page express the same, that volset values are there to label a group of CDs so machines can know that kind of info by just reading bits from the CDs, not expecting someone reads the feltmarker text and types it. And if the OS does not support that kind of info or is unable to do something with it (ie allow apps to check that next CD is really next CD), I would expect kernel to shut up and ignore it, not take rare measures. So do I not understand the man page and doc? Are they / is one wrong? Or is the Linux kernel wrong? Do I just upgrade something to fix it or I have to contact someone and hope he fixes the problem (man page or src)? If kernel problem (my guess) the man page could mention that the option can cause problems even if the usage is right. Thanks in advance. GSR -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
