rwood@ccs.carleton.ca (Richard Wood) (10/12/90)
Is anyone out there familiar with rbak? I am trying to restore
directory /local to an apollo SR9.704 system using the command
rbak -dev ct1 -f 14 //a/local
Rbak successfully recovers those /local subdirectories and files
stored at the beginning of tape-file 14, but about halfway through it
aborts with the message
?(rbak) (heap_$get) Heap table full.
We have tried rebooting, using a different backup tape, logging on
under different accounts, asking only for specific subdirectories from
the second half of file 14, increasing the free disc space and free
memory, etc, but still we can't get past that halfway point.
Rbak has no problems, however, producing an index of file 14 - the
stuff is all there.
Does anyone know what is happening here? Could it be lack of free
memory or disc space (in spite of the changes we tried)? Is there any
way to correct the problem? To get around it?
Thanks in advance.....
--
Richard Wood rwood@ccs.carleton.cadbfunk@ICAEN.UIOWA.EDU (David B Funk) (10/12/90)
IN posting <1990Oct11.182457.14872@ccs.carleton.ca>, Richard Wood asks: > Is anyone out there familiar with rbak? I am trying to restore > directory /local to an apollo SR9.704 system using the command > rbak -dev ct1 -f 14 //a/local > Rbak successfully recovers those /local subdirectories and files > stored at the beginning of tape-file 14, but about halfway through it > aborts with the message > ?(rbak) (heap_$get) Heap table full. There was a bug (limitation?) in the FCS sr9.7 rbak that could cause it to die with this error, particularly if the tape had lots of ACL objects on it. There is a sr9.7 patch (Patch 136 New rbak for 9.7) to fix it. If you have any sr10 systems around, check out the rbak that lives in the sr9.7_compatibility directory, it may work. (/sr9.7_compatibility/sr9.7_executables/com/rbak). If you don't already have a reasonably current sr9.7 patch tape, get one ASAP. HPollo is threatening to drop support for sr9.7 in '91. Dave Funk
system@alchemy.chem.utoronto.ca (System Admin (Mike Peterson)) (10/20/90)
In article <1990Oct11.182457.14872@ccs.carleton.ca> rwood@ccs.carleton.ca (Richard Wood) writes: >Is anyone out there familiar with rbak? I am trying to restore >directory /local to an apollo SR9.704 system using the command > rbak -dev ct1 -f 14 //a/local > [ stuff deleted ] There is a problem with rbak on DN10000 nodes when a component of the path is only a single character - it fails with a segmentation fault part way through the restore. Apollo says this only exists on 10K's, but maybe not ?! -- Mike Peterson, System Administrator, U/Toronto Department of Chemistry E-mail: system@alchemy.chem.utoronto.ca Tel: (416) 978-7094 Fax: (416) 978-8775