Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!bonnie.concordia.ca!thunder.mcrcim.mcgill.edu!snorkelwacker.mit.edu!bu.edu!rpi!zaphod!uakari.primate.wisc.edu!unmvax!bbx!tantalum!cheeks From: cheeks@edsr.eds.com (Mark Costlow) Newsgroups: comp.databases Subject: Sybase gripes Message-ID: <5716@tantalum.UUCP> Date: 15 Jan 91 00:17:09 GMT Sender: usenet@tantalum.UUCP Reply-To: cheeks@edsr.eds.com Organization: EDS Research Lines: 43 Hi. I'm in charge of administering a moderately large sybase database, and I've run into a couple of things that bother me. So, I thought I'd whine a little and see if anyone else has worked around these problems: 1. Remote tape drives. In this age of networked computers, and considering all of the network-based functionality that sybase DOES have, not supporting remote archive devices seem completely ludicrous. Has anywone done anything to work around this problem? (I was thinking one could set up a FIFO as a "disk byte stream" dump device and have a daemon ship the data to a remote tape, but I haven't had time to play with anything like that yet). 2. Lack of support for cartridge tapes. I can't seem to find any way to get sybase to exercise any kind of tape control on a scsi cartridge tape device. In fact, to get sybase to write to the device at all, you have to tell it that it's a disk! That means I can't have a database any larger than my largest tape device. This also strikes me as silly! Do I really have to buy a 1/2" mag tape drive to dump my database, when my exabyte would be so much more efficient? 3. More lack of support for cartridge tapes This is related to #2. I'm running on a Sun-4/470 with all SCSI disks, under SunOS 4.1.1. I have my exabyte set up as a disk byte stream device. On about 1/2 of the dumps I do, I get a "st?: tape synchronisation lost" message from the OS (where ? is replaced by the particular tape device). I assume that sybase is doing (or not doing) something to the tape device to cause this, but I haven't got a clue what. The tape drive works fine for normal dumps and tars. Has anyone else seen this? Barring a solution for #2, does anyone have any ideas that I could try? Any input will be appreciated. Thanks, Mark -- cheeks@edsr.eds.com or ...uunet!edsr!cheeks