kka059@MIPL3.JPL.NASA.GOV (07/21/87)
Recently, we observed a curious behaviour on our system. One of our disks had become so fragmented/full that anybody that attempted to create a file of any size greater than a few blocks received either a "ACP-E-CRE, file create failed" message or a "RMS-E-CRE, ACP file creation failed, %SYSTEM-W-HEADERFULL, ..." error. The curious behaviour was that if someone created a small file, then successively extended the file using SYS$EXTEND, the system would bugcheck (and crash of course)! Although I would think that it would be intuitively obvious, we would rather have an error condition returned than to have a problem on one disk (out of a farm of more than 25 disks) crash the system. Since rebuilding the offending disk via backup/restore, the problem has not been repeated. Has anyone else experienced this problem? Has anyone SPR'd it????? +----------------------------+------------------------------------------+ | Kurt Andersen | Jet Propulsion Laboratory | | MIPL Applications | Mail Stop 168-427 | | Programmer | 4800 Oak Grove Drive | | Office: 169-425 | Pasadena, Calif. 91109 | +----------------------------+-------------------------+----------------+ | NETWORKS: | Eat, drink and | | SPAN: Mipl3::KKA059 (5.153) | be merry... | | ARPA Internet: KKA059@Mipl3.Jpl.Nasa.Gov | | | Internet Address: [128.149.1.28] | | | ARPAnet->SPAN: KKA059%Mipl3@Su-Star.Stanford.Edu | for tomorrow | | or: KKA059%Mipl3@Jpl-Vlsi.Arpa | the system | | Ma Bell (R.I.P.): (818) 354-1672 | will prang. | +------------------------------------------------------+----------------+ | Look it up... | +-----------------------------------------------------------------------+