GET BACK TO ME ASAP.
Good day, I am Mr. Song Lile, staff of Hang Seng Bank Hong Kong,I have a business proposal for you, contact Email:mrsonglil...@yahoo.com.hk Regards, Song Lile -- To UNSUBSCRIBE, email to cdwrite-requ...@other.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@other.debian.org
Re: Announcing xorriso-0.3.6
Thomas Schmitt scdbac...@gmx.net wrote: All recent software uses ACLs based on NTFS ACLs. Not according to the Linux and FreeBSD man pages which i studied. If i ever get in touch with other ACL permissions like Modify, List Folder or Full Control then i will use the FUTURE_VERSION type of the current definition to define a 2-byte entry format. It seems like you missunderstand things. NTFS-style ACLs have been standardized in 2004 together with NFSv4. The UFS-style ACLs you like to implement have been withdrawn in 1999, they only continue to exist for legacy reasons. I know of no future ACL style, but about what people currently use and this is NTFS-style ACLs. If you don't believe me, check the ACL implementation that is supported on NetApps servers. BTW: do you know of any OS that reads your ACL implementation? Not yet. A strong hint for not adding support to Rock Ridge for something that has been withdrawn 10 years ago... Note that regarding to extended attributes, we are in a similar situation. The implementation from you is incompatible to the current standard for extended attributes that is is the NFSv4 standard. The related basics have been discussed in Y2000 on the POSIX mailing list and the related basic system interface was introduced in Solaris in August 2000 and later added to POSIX.2-2008. BTW: Linux and FreeBSD implement NFSv4 and with NFSv4 at least the extended attribute standard from NFSv4. I am sure that these OS will also implement NFSv4 style ACLs is this was not yet done already. It seems that your proposal is 10 years too late Jörg -- EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin j...@cs.tu-berlin.de(uni) joerg.schill...@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily -- To UNSUBSCRIBE, email to cdwrite-requ...@other.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@other.debian.org
Re: Announcing xorriso-0.3.6
Hi, The UFS-style ACLs you like to implement have been withdrawn I implemented man 5 acl of Linux. BTW: Linux and FreeBSD implement NFSv4 and with NFSv4 at least the extended attribute standard from NFSv4. On FreeBSD i found no trace of xattr and on Linux i found man 5 attr. AAIP could attach 8 TB of attribute to a file. In the isofs. namespace one could define a link attribute which connects files to files. So if interested Solaris users would provide testing and documentation then one could surely find a nice reversible mapping of Solaris xattr to AAIP. Elsewise i will wait until such features appear in Linux. Have a nice day :) Thomas -- To UNSUBSCRIBE, email to cdwrite-requ...@other.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@other.debian.org
Re: Announcing xorriso-0.3.6
Thomas Schmitt scdbac...@gmx.net wrote: The UFS-style ACLs you like to implement have been withdrawn I implemented man 5 acl of Linux. This is the re-implementation of the UFS-style ACL implementation from Solaris that appeared on Solaris in 1994 and that was discussed for some years in the POSIX standard commitee. It was withdrawn in 1999. UFS-style ACLs currently only exist for backward compatibility. UFS-style ACLs do not include support for deny ACL entries and for this reason, users did not like UFS-style ACLs. The CIFS server on Solaris can directly use the ACLs from ZFS, on Linux samba needs to implement an ACL simulation as there is no compatible ACL support. Modern OS are currently going towards ZFS (Mac OS X is currently porting - FreeBSD did already port ZFS). The fact that two basic features (NFS and ZFS) use the current ACL standard makes it obvious that the time of the old UFS-style ACLs is over. BTW: Linux and FreeBSD implement NFSv4 and with NFSv4 at least the extended attribute standard from NFSv4. On FreeBSD i found no trace of xattr and on Linux i found man 5 attr. As FreeBSD implements ZFS, there is a big chance that FreeBSD will adopt the extended attributes from Solaris. This implementation is heavily used by the DOS type attributes from ZFS (system flag, archive flag, ...) and AFAIK for the DOS short name that is supported by ZFS. ZFS allows to attach zetabytes of attribute space to a file. AAIP could attach 8 TB of attribute to a file. In the isofs. namespace one could define a link attribute which connects files to files. I am not sure what you are talking about... Mkisofs implements support for hardlinks on Rock Ridge and Solaris does the same. If you like to see hard links on Linux, I recommend just to implement support for inode numbers on Linux So if interested Solaris users would provide testing and documentation then one could surely find a nice reversible mapping of Solaris xattr to AAIP. Elsewise i will wait until such features appear in Linux. Xattr features exist on Solaris, on Solaris they are compliant to the NFSv4 spec. If Linux users are interested on the current ACL standard, I recommend to add ZFS to the basic filesystems supported by Linux. Jörg -- EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin j...@cs.tu-berlin.de(uni) joerg.schill...@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily -- To UNSUBSCRIBE, email to cdwrite-requ...@other.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@other.debian.org
Re: Announcing xorriso-0.3.6
Thomas Schmitt scdbac...@gmx.net wrote: If you like to see hard links on Linux, I recommend just to implement support for inode numbers on Linux My trick is that i implement such things in xorriso. My influence on kernels is limited. But xorriso can retrieve any extra information that it encoded in the image. My trick is to implement standards in the Solaris kernel to get support for standard compliant features in mkisofs ;-) If Linux users are interested on the current ACL standard, I recommend to add ZFS to the basic filesystems supported by Linux. Seems that those who could are not interested and those who would are not capable. I would call it NIH Syndrom :-( It can be overcome if enough people ask for support. As said, i have no influence on kernels. And nobody has influence on the installed systems. So i stay with what i can control: xorriso. The best I can do is to implement things in the kernel and to convince people to do more collaboration. Unfortunately the Linux kernel people are not interested in kernel extensions. On Solaris, it is easier to get things into the kernel if you clearly explain things and in case things fit. Convincing people is harder... but I convinced people that GPL code may call CDDL code. Now it is only a matter of interest in the Linux camp for ZFS to appear inside Linux. Jörg -- EMail:jo...@schily.isdn.cs.tu-berlin.de (home) Jörg Schilling D-13353 Berlin j...@cs.tu-berlin.de(uni) joerg.schill...@fokus.fraunhofer.de (work) Blog: http://schily.blogspot.com/ URL: http://cdrecord.berlios.de/private/ ftp://ftp.berlios.de/pub/schily -- To UNSUBSCRIBE, email to cdwrite-requ...@other.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@other.debian.org
Re: Announcing xorriso-0.3.6
Thomas Schmitt wrote: Hi, If you like to see hard links on Linux, I recommend just to implement support for inode numbers on Linux My trick is that i implement such things in xorriso. My influence on kernels is limited. But xorriso can retrieve any extra information that it encoded in the image. It is also able to restore multi extent files on operating systems which fail to implement a proper reader for ISO level 3 (quite new Linux kernels among them). If Linux users are interested on the current ACL standard, I recommend to add ZFS to the basic filesystems supported by Linux. Seems that those who could are not interested and those who would are not capable. As long as you can back up and restore what is there in Linux now, and use features which other applications and OS are required to ignore if they do not understand, then you have provided a useful backup capability which is portable enough to be useful. Rarely are the xattr needed to be moved to other OS, because the exact functionality of the xattr may not be precisely the same. If the file nuances are preserved within the same OS that's sufficient to be useful. If IBM actually does buy Sun, I suspect that the best features of AIX and Solaris will be made available to Linux, and marketing of the legacy OS will be minimal. For the same reasons that tru64, MULTICS and OS/2 are no longer competing, it's no longer a question of which OS is best (by whatever measure), it's a measure of which OS generates more revenue than it costs to maintain. -- bill davidsen david...@tmr.com CTO TMR Associates, Inc You are disgraced professional losers. And by the way, give us our money back. - Representative Earl Pomeroy, Democrat of North Dakota on the A.I.G. executives who were paid bonuses after a federal bailout.
