kdq@pthya.UUCP ( Kip Quackenbush) (06/26/84)
<<eat me>> Is there a method of determining a file's block # under SYS V? Problem: Took many device errors on a particular block. Would like to find out the file(s) that reside in that block. My guess is that once you know the filename, you can copy it, then move the original to outer space, and bingo, the potential read of the bad block will not happen. Solution?..................... -- Kip Quackenbush pthya!kdq {ihnp4,ucbvax,cbosgd,decwrl,amd70,fortune,zehntel}!dual!pthya!kdq Pacific Bell, San Francisco, California
dave@uwvax.ARPA (06/29/84)
There should be holdovers on Sys V from version 7, namely icheck and ncheck (names may have changed -- its been a while). Method: icheck -b <blocknumbers> <filesystem> this gets you the inode containing the specified block(s) ncheck -i <i-numbers> <filesystem> this gets you the filename(s) They are both described in the manual: icheck(8), ncheck(8). -- Dave Cohrs @ wisconsin ...!{allegra,heurikon,ihnp4,seismo,sfwin,ucbvax,uwm-evax}!uwvax!dave dave@wisc-rsch.arpa
ron@BRL-TGR.ARPA (06/29/84)
From: Ron Natalie <ron@BRL-TGR.ARPA> We used to do this with icheck -b (tell me what inodes this block exists in) and ncheck -i (give me the name of this inode). Unfortunately icheck was replaced by fsck and fsck doesn't seem to have this feature. -Ron
fair@dual.UUCP (Erik E. Fair) (07/04/84)
What you need to do, Kip, is get the logical block number first. If all you have is the physical block number, you have to figure out which filesystem it's part of (i.e. which logical device). Once you do that you can do this: % icheck -b bn# bn# bn# ... /dev/rhp0a inode xxxx % ncheck -i xxxx /dev/rhp0a xxxxx /foo/bar/bletch Icheck will waltz through the inodes looking for the inodes which claim your bad block. When you have the inode number (or numbers in the unfortunate case of dups), you can get the names which reference those inodes from ncheck, which waltzes down the directories to this end. Clear? Erik E. Fair ucbvax!fair fair@ucb-arpa.ARPA dual!fair@Berkeley.ARPA {ihnp4,ucbvax,cbosgd,decwrl,amd70,fortune,zehntel}!dual!fair Dual Systems Corporation, Berkeley, California
jm@wlbr.UUCP (07/09/84)
Unfortunately, icheck(1M) is not distributed with System V. However, ncheck is, and may easily be hacked to yield a superior substitue. Jim Macropol {ihnp4,scgvaxd,trwrb,vortex}!wlbr!jm