Notes from when I looked into unzip on June 11, 2004. --------------------------------------------------------------------------- I was trying to fix the unzip errors we've been getting from the WO DVD's for the last year. For example, as root on patimg2, doing a unzip /cdrom/impact242004_1/US03/012095/WO2003_089902_10062004.zip \ -d /images/pamscan/impact242004_1/03-089902 command, would give these errors, Archive: /cdrom/impact242004_1/US03/012095/WO2003_089902_10062004.zip End-of-central-directory signature not found. Either this file is not a zipfile, or it constitutes one disk of a multi-part archive. In the latter case the central directory and zipfile comment will be found on the last disk(s) of this archive. note: /cdrom/impact242004_1/US03/012095/WO2003_089902_10062004.zip may be a plain executable, not an archive unzip: cannot find zipfile directory in one of /cdrom/impact242004_1/US03/012095/WO2003_089902_10062004.zip or /cdrom/impact242004_1/US03/012095/WO2003_089902_10062004.zip.zip, and cannot find /cdrom/impact242004_1/US03/012095/WO2003_089902_10062004.zip.ZIP, period. We were using the unzip from /local/bin, which gave these lines doing a unzip -v command UnZip 5.50 of 17 February 2002, by Info-ZIP. Maintained by C. Spieler. Send bug reports to the authors at Zip-Bugs@lists.wku.edu; see README for details. Latest sources and executables are at ftp://ftp.info-zip.org/pub/infozip/ ; see ftp://ftp.info-zip.org/pub/infozip/UnZip.html for other sites. ... --------------------------------------------------------------------------- ftp'd unzip551.tar.gz from ftp.info-zip.org/pub/infozip/src gzip -d < unzip551.tar.gz | tar xf - mv unzip-5.51 /afs/d/software/base cd /afs/d/software/base/unzip-5.51 cp -p unix/Makefile . To get my feet wet, I did make make list To finally do the make, I make generic Besides a bunch of 1500-030: (I) INFORMATION: inflate_fixed: Additional optimization may be attained by recompiling and specifying MAXMEM option with a value greater than 2048. messages, I got this funky error message, ... rm -f process_.c ln ttyio.c ttyio_.c cc -c -O -I. -DUNIX -DSFX ttyio_.c rm -f ttyio_.c ln unix/unix.c unix_.c ln: 0653-422 unix/unix.c and unix_.c are located on different file systems. make: 1254-004 The error code from the last command is 1. make: 1254-005 Ignored error code 1 from last command. cc -c -O -I. -DUNIX -DSFX unix_.c cc: 1501-228 input file unix_.c not found make: 1254-004 The error code from the last command is 252. Stop. Is this an AFS thing? Tried this in a local file system on jasper, cd /tmp gzip -d < /u/jasper/unzip551.tar.gz | tar xf - cd /tmp/unzip-5.51 cp -p unix/Makefile . make generic And it worked fine. Great. It looks like the "make install" will install in /usr/local/bin and /usr/local/man. Ok, gotta be root first. su - cd /tmp/unzip-5.51 make install Showed me that it did this mkdir -p /usr/local/bin cp unzip funzip unzipsfx /usr/local/bin cp unix/zipgrep /usr/local/bin rm -f /usr/local/bin/zipinfo ln /usr/local/bin/unzip /usr/local/bin/zipinfo mkdir -p /usr/local/man/man1 cp man/funzip.1 /usr/local/man/man1/funzip.1 cp man/unzip.1 /usr/local/man/man1/unzip.1 cp man/unzipsfx.1 /usr/local/man/man1/unzipsfx.1 cp man/zipgrep.1 /usr/local/man/man1/zipgrep.1 cp man/zipinfo.1 /usr/local/man/man1/zipinfo.1 chmod 755 /usr/local/bin/funzip /usr/local/bin/unzip \ /usr/local/bin/unzipsfx /usr/local/bin/zipgrep \ /usr/local/bin/zipinfo chmod 644 /usr/local/man/man1/funzip.1 /usr/local/man/man1/unzip.1 \ /usr/local/man/man1/unzipsfx.1 /usr/local/man/man1/zipgrep.1 \ /usr/local/man/man1/zipinfo.1 To move this back into AFS (as jasper again so I have the right tokens), cd /tmp mv unzip-5.51 /afs/d/software/base/unzip-5.51 cd /afs/d/software/base/unzip-5.51 To install into AFS's /local/bin instead, make prefix=/local install --------------------------------------------------------------------------- Now a unzip -v command shows me UnZip 5.51 of 22 May 2004, by Info-ZIP. ... but I still get the same error with the above unzip command. So getting the latest version didn't fix anything. --------------------------------------------------------------------------- Could it have anything to do with the zone we're in? That's important for movie DVD's. Is it for data DVD's, too? --------------------------------------------------------------------------- Got a couple of notes from Stephan Pichlmaier from Munich, when Sandy and/or Rebecca mailed him one of our WO DVD's (impact102004) where he says that the bytes when reading it from one of our systems, versus one of his, are different. Just comparing bytes for one file, saw byte differences starting at x'1FE00' = 130,560 bytes. Why would that be? How could that happen? --------------------------------------------------------------------------- Looking at impact222004 (real label=IB1274_R87_-250(1)_DVD_1): From the PC, * G:\US99\012269\WO1999_062765_27052004.zip = 250,448 bytes * Power Archiver can unzip it just fine. * Copied to AFS, AFS's unzip works just fine on both jasper & patimg2. From patimg2, * /cdrom/impact222004/US99/012269/WO1999_062765_27052004.zip = 250,448 bytes * But unzip-ing it from patimg2, gives errors, unzip /cdrom/impact222004/US99/012269/WO1999_062765_27052004.zip -d /images/pamscan/impact222004/99-062765 Archive: /cdrom/impact222004/US99/012269/WO1999_062765_27052004.zip End-of-central-directory signature not found. Either this file is not a zipfile, or it constitutes one disk of a multi-part archive. In the latter case the central directory and zipfile comment will be found on the last disk(s) of this archive. note: /cdrom/impact222004/US99/012269/WO1999_062765_27052004.zip may be a plain executable, not an archive unzip: cannot find zipfile directory in one of /cdrom/impact222004/US99/012269/WO1999_062765_27052004.zip or /cdrom/impact222004/US99/012269/WO1999_062765_27052004.zip.zip, and cannot find /cdrom/impact222004/US99/012269/WO1999_062765_27052004.zip.ZIP, period. Running the zip file through od and comparing it against the version read from Windows, shows the last 80 bytes wrong. Even more, the last 80 bytes are repeated from 512 bytes ago, od -x /cdrom/impact222004/US99/012269/WO1999_062765_27052004.zip | tail -38 Decimal Octal =========== Data ==================== 249856 0750000 0000 0000 0000 0000 0000 0000 0932 0100 <- This data 249872 0750020 3030 3030 3032 2e74 6966 504b 0102 1400 <- is 249888 0750040 1400 0800 0800 6455 ae30 1eca aec7 71cf <- duplicated 249904 0750060 0000 22da 0000 0a00 0000 0000 0000 0000 <- below 249920 0750100 0000 0000 0f42 0100 3030 3030 3033 2e74 <- 249936 0750120 6966 504b 0102 1400 1400 0800 0800 6455 249952 0750140 ae30 31ec c071 1b3c 0000 3045 0000 0a00 ... 250352 0750760 7469 6f6e 2e78 6d6c 504b 0102 1400 1400 250368 0751000 0000 0000 0000 0000 0000 0000 0932 0100 <- This data is 250384 0751020 3030 3030 3032 2e74 6966 504b 0102 1400 <- wrong and is 250400 0751040 1400 0800 0800 6455 ae30 1eca aec7 71cf <- duplicated 250416 0751060 0000 22da 0000 0a00 0000 0000 0000 0000 <- from above. 250432 0751100 0000 0000 0f42 0100 3030 3030 3033 2e74 <- The bottom line is, this is obviously an iXos/jukeman software bug. --------------------------------------------------------------------------- Since I last looked, the iXos/jukeman software (see my aixnotes/image file) was previously owned by Legato, but now EMC owns it. I decided instead to try to build up a Linux machine to drive the jukebox. See my aixnotes/fsx file for those details. ... Five weeks later, I've confirmed that Linux can read it just fine.