1 .\" Copyright (c) 1982, 1993
2 .\" The Regents of the University of California. All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
7 .\" 1. Redistributions of source code must retain the above copyright
8 .\" notice, this list of conditions and the following disclaimer.
9 .\" 2. Redistributions in binary form must reproduce the above copyright
10 .\" notice, this list of conditions and the following disclaimer in the
11 .\" documentation and/or other materials provided with the distribution.
12 .\" 3. All advertising materials mentioning features or use of this software
13 .\" must display the following acknowledgement:
14 .\" This product includes software developed by the University of
15 .\" California, Berkeley and its contributors.
16 .\" 4. Neither the name of the University nor the names of its contributors
17 .\" may be used to endorse or promote products derived from this software
18 .\" without specific prior written permission.
20 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
32 .\" @(#)4.t 8.1 (Berkeley) 6/5/93
34 .ds RH Appendix A \- Fsck Error Conditions
36 Appendix A \- Fsck Error Conditions
42 a multi-pass file system check program.
43 Each file system pass invokes a different Phase of the
46 After the initial setup,
48 performs successive Phases over each file system,
49 checking blocks and sizes,
53 and the map of free blocks,
54 (possibly rebuilding it),
55 and performs some cleanup.
59 is run non-interactively to
61 the file systems after an unclean halt.
62 While preen'ing a file system,
63 it will only fix corruptions that are expected
64 to occur from an unclean halt.
65 These actions are a proper subset of the actions that
67 will take when it is running interactively.
68 Throughout this appendix many errors have several options
69 that the operator can take.
70 When an inconsistency is detected,
72 reports the error condition to the operator.
73 If a response is required,
75 prints a prompt message and
77 When preen'ing most errors are fatal.
78 For those that are expected,
79 the response taken is noted.
80 This appendix explains the meaning of each error condition,
81 the possible responses, and the related error conditions.
83 The error conditions are organized by the
87 program in which they can occur.
88 The error conditions that may occur
89 in more than one Phase
90 will be discussed in initialization.
94 Before a file system check can be performed, certain
95 tables have to be set up and certain files opened.
96 This section concerns itself with the opening of files and
97 the initialization of tables.
98 This section lists error conditions resulting from
103 file system size checks,
104 and creation of the scratch file.
105 All the initialization errors are fatal
106 when the file system is being preen'ed.
111 \fIC\fP is not a legal option to
113 legal options are \-b, \-c, \-y, \-n, and \-p.
115 terminates on this error condition.
118 manual entry for further detail.
121 .B "cannot alloc NNN bytes for blockmap"
123 .B "cannot alloc NNN bytes for freemap"
125 .B "cannot alloc NNN bytes for statemap"
127 .B "cannot alloc NNN bytes for lncntp"
130 request for memory for its virtual
131 memory tables failed.
132 This should never happen.
134 terminates on this error condition.
138 .B "Can't open checklist file: \fIF\fP"
140 The file system checklist file
143 can not be opened for reading.
145 terminates on this error condition.
146 Check access modes of \fIF\fP.
152 request for statistics about the root directory ``/'' failed.
153 This should never happen.
155 terminates on this error condition.
159 .B "Can't stat \fIF\fP"
161 .B "Can't make sense out of name \fIF\fP"
164 request for statistics about the file system \fIF\fP failed.
165 When running manually,
166 it ignores this file system
167 and continues checking the next file system given.
168 Check access modes of \fIF\fP.
171 .B "Can't open \fIF\fP"
174 request attempt to open the file system \fIF\fP failed.
175 When running manually, it ignores this file system
176 and continues checking the next file system given.
177 Check access modes of \fIF\fP.
180 .B "\fIF\fP: (NO WRITE)"
182 Either the \-n flag was specified or
184 attempt to open the file system \fIF\fP for writing failed.
185 When running manually,
186 all the diagnostics are printed out,
187 but no modifications are attempted to fix them.
190 .B "file is not a block or character device; OK"
194 a regular file name by mistake.
195 Check the type of the file specified.
197 Possible responses to the OK prompt are:
199 ignore this error condition.
201 ignore this file system and continues checking
202 the next file system given.
205 .B "UNDEFINED OPTIMIZATION IN SUPERBLOCK (SET TO DEFAULT)"
207 The superblock optimization parameter is neither OPT_TIME
210 Possible responses to the SET TO DEFAULT prompt are:
212 The superblock is set to request optimization to minimize
213 running time of the system.
214 (If optimization to minimize disk space utilization is
215 desired, it can be set using \fItunefs\fP(8).)
217 ignore this error condition.
220 .B "IMPOSSIBLE MINFREE=\fID\fP IN SUPERBLOCK (SET TO DEFAULT)"
222 The superblock minimum space percentage is greater than 99%
225 Possible responses to the SET TO DEFAULT prompt are:
227 The minfree parameter is set to 10%.
228 (If some other percentage is desired,
229 it can be set using \fItunefs\fP(8).)
231 ignore this error condition.
234 .B "IMPOSSIBLE INTERLEAVE=\fID\fP IN SUPERBLOCK (SET TO DEFAULT)"
236 The file system interleave is less than or equal to zero.
238 Possible responses to the SET TO DEFAULT prompt are:
240 The interleave parameter is set to 1.
242 ignore this error condition.
245 .B "IMPOSSIBLE NPSECT=\fID\fP IN SUPERBLOCK (SET TO DEFAULT)"
247 The number of physical sectors per track is less than the number
248 of usable sectors per track.
250 Possible responses to the SET TO DEFAULT prompt are:
252 The npsect parameter is set to the number of usable sectors per track.
254 ignore this error condition.
257 One of the following messages will appear:
259 .B "MAGIC NUMBER WRONG"
261 .B "NCG OUT OF RANGE"
263 .B "CPG OUT OF RANGE"
265 .B "NCYL DOES NOT JIVE WITH NCG*CPG"
267 .B "SIZE PREPOSTEROUSLY LARGE"
269 .B "TRASHED VALUES IN SUPER BLOCK"
271 and will be followed by the message:
273 .B "\fIF\fP: BAD SUPER BLOCK: \fIB\fP"
275 .B "USE -b OPTION TO FSCK TO SPECIFY LOCATION OF AN ALTERNATE"
277 .B "SUPER-BLOCK TO SUPPLY NEEDED INFORMATION; SEE fsck(8)."
279 The super block has been corrupted.
280 An alternative super block must be selected from among those
283 (8) when the file system was created.
284 For file systems with a blocksize less than 32K,
285 specifying \-b 32 is a good first choice.
288 .B "INTERNAL INCONSISTENCY: \fIM\fP"
291 has had an internal panic, whose message is specified as \fIM\fP.
292 This should never happen.
296 .B "CAN NOT SEEK: BLK \fIB\fP (CONTINUE)"
299 request for moving to a specified block number \fIB\fP in
300 the file system failed.
301 This should never happen.
304 Possible responses to the CONTINUE prompt are:
306 attempt to continue to run the file system check.
308 however the problem will persist.
309 This error condition will not allow a complete check of the file system.
312 should be made to re-check this file system.
313 If the block was part of the virtual memory buffer
316 will terminate with the message ``Fatal I/O error''.
318 terminate the program.
321 .B "CAN NOT READ: BLK \fIB\fP (CONTINUE)"
324 request for reading a specified block number \fIB\fP in
325 the file system failed.
326 This should never happen.
329 Possible responses to the CONTINUE prompt are:
331 attempt to continue to run the file system check.
332 It will retry the read and print out the message:
334 .B "THE FOLLOWING SECTORS COULD NOT BE READ: \fIN\fP"
336 where \fIN\fP indicates the sectors that could not be read.
339 ever tries to write back one of the blocks on which the read failed
340 it will print the message:
342 .B "WRITING ZERO'ED BLOCK \fIN\fP TO DISK"
344 where \fIN\fP indicates the sector that was written with zero's.
345 If the disk is experiencing hardware problems, the problem will persist.
346 This error condition will not allow a complete check of the file system.
349 should be made to re-check this file system.
350 If the block was part of the virtual memory buffer
353 will terminate with the message ``Fatal I/O error''.
355 terminate the program.
358 .B "CAN NOT WRITE: BLK \fIB\fP (CONTINUE)"
361 request for writing a specified block number \fIB\fP
362 in the file system failed.
363 The disk is write-protected;
364 check the write protect lock on the drive.
365 If that is not the problem, see a guru.
367 Possible responses to the CONTINUE prompt are:
369 attempt to continue to run the file system check.
370 The write operation will be retried with the failed blocks
371 indicated by the message:
373 .B "THE FOLLOWING SECTORS COULD NOT BE WRITTEN: \fIN\fP"
375 where \fIN\fP indicates the sectors that could not be written.
376 If the disk is experiencing hardware problems, the problem will persist.
377 This error condition will not allow a complete check of the file system.
380 should be made to re-check this file system.
381 If the block was part of the virtual memory buffer
384 will terminate with the message ``Fatal I/O error''.
386 terminate the program.
389 .B "bad inode number DDD to ginode"
391 An internal error has attempted to read non-existent inode \fIDDD\fP.
397 Phase 1 \- Check Blocks and Sizes
399 This phase concerns itself with
401 This section lists error conditions resulting from
402 checking inode types,
403 setting up the zero-link-count table,
404 examining inode block numbers for bad or duplicate blocks,
406 and checking inode format.
407 All errors in this phase except
408 .B "INCORRECT BLOCK COUNT"
410 .B "PARTIALLY TRUNCATED INODE"
411 are fatal if the file system is being preen'ed.
414 .B "UNKNOWN FILE TYPE I=\fII\fP (CLEAR)"
416 The mode word of the inode \fII\fP indicates that the inode is not a
417 special block inode, special character inode, socket inode, regular inode,
418 symbolic link, or directory inode.
420 Possible responses to the CLEAR prompt are:
422 de-allocate inode \fII\fP by zeroing its contents.
423 This will always invoke the UNALLOCATED error condition in Phase 2
424 for each directory entry pointing to this inode.
426 ignore this error condition.
429 .B "PARTIALLY TRUNCATED INODE I=\fII\fP (SALVAGE)"
432 has found inode \fII\fP whose size is shorter than the number of
433 blocks allocated to it.
434 This condition should only occur if the system crashes while in the
435 midst of truncating a file.
436 When preen'ing the file system,
438 completes the truncation to the specified size.
440 Possible responses to SALVAGE are:
442 complete the truncation to the size specified in the inode.
444 ignore this error condition.
447 .B "LINK COUNT TABLE OVERFLOW (CONTINUE)"
449 An internal table for
451 containing allocated inodes with a link count of
452 zero cannot allocate more memory.
453 Increase the virtual memory for
456 Possible responses to the CONTINUE prompt are:
458 continue with the program.
459 This error condition will not allow a complete check of the file system.
462 should be made to re-check this file system.
463 If another allocated inode with a zero link count is found,
464 this error condition is repeated.
466 terminate the program.
469 .B "\fIB\fP BAD I=\fII\fP"
471 Inode \fII\fP contains block number \fIB\fP with a number
472 lower than the number of the first data block in the file system or
473 greater than the number of the last block
475 This error condition may invoke the
476 .B "EXCESSIVE BAD BLKS"
477 error condition in Phase 1 (see next paragraph) if
478 inode \fII\fP has too many block numbers outside the file system range.
479 This error condition will always invoke the
481 error condition in Phase 2 and Phase 4.
484 .B "EXCESSIVE BAD BLKS I=\fII\fP (CONTINUE)"
486 There is more than a tolerable number (usually 10) of blocks with a number
487 lower than the number of the first data block in the file system or greater than
488 the number of last block in the file system associated with inode \fII\fP.
490 Possible responses to the CONTINUE prompt are:
492 ignore the rest of the blocks in this inode
493 and continue checking with the next inode in the file system.
494 This error condition will not allow a complete check of the file system.
497 should be made to re-check this file system.
499 terminate the program.
502 .B "BAD STATE DDD TO BLKERR"
504 An internal error has scrambled
506 state map to have the impossible value \fIDDD\fP.
512 .B "\fIB\fP DUP I=\fII\fP"
514 Inode \fII\fP contains block number \fIB\fP that is already claimed by
516 This error condition may invoke the
517 .B "EXCESSIVE DUP BLKS"
518 error condition in Phase 1 if
519 inode \fII\fP has too many block numbers claimed by other inodes.
520 This error condition will always invoke Phase 1b and the
522 error condition in Phase 2 and Phase 4.
525 .B "EXCESSIVE DUP BLKS I=\fII\fP (CONTINUE)"
527 There is more than a tolerable number (usually 10) of blocks claimed by other
530 Possible responses to the CONTINUE prompt are:
532 ignore the rest of the blocks in this inode
533 and continue checking with the next inode in the file system.
534 This error condition will not allow a complete check of the file system.
537 should be made to re-check this file system.
539 terminate the program.
542 .B "DUP TABLE OVERFLOW (CONTINUE)"
546 containing duplicate block numbers cannot allocate any more space.
547 Increase the amount of virtual memory available to
550 Possible responses to the CONTINUE prompt are:
552 continue with the program.
553 This error condition will not allow a complete check of the file system.
556 should be made to re-check this file system.
557 If another duplicate block is found, this error condition will repeat.
559 terminate the program.
562 .B "PARTIALLY ALLOCATED INODE I=\fII\fP (CLEAR)"
564 Inode \fII\fP is neither allocated nor unallocated.
566 Possible responses to the CLEAR prompt are:
568 de-allocate inode \fII\fP by zeroing its contents.
570 ignore this error condition.
573 .B "INCORRECT BLOCK COUNT I=\fII\fP (\fIX\fP should be \fIY\fP) (CORRECT)"
575 The block count for inode \fII\fP is \fIX\fP blocks,
576 but should be \fIY\fP blocks.
577 When preen'ing the count is corrected.
579 Possible responses to the CORRECT prompt are:
581 replace the block count of inode \fII\fP with \fIY\fP.
583 ignore this error condition.
585 Phase 1B: Rescan for More Dups
587 When a duplicate block is found in the file system, the file system is
588 rescanned to find the inode that previously claimed that block.
589 This section lists the error condition when the duplicate block is found.
592 .B "\fIB\fP DUP I=\fII\fP"
594 Inode \fII\fP contains block number \fIB\fP that
595 is already claimed by another inode.
596 This error condition will always invoke the
598 error condition in Phase 2.
599 You can determine which inodes have overlapping blocks by examining
600 this error condition and the DUP error condition in Phase 1.
602 Phase 2 \- Check Pathnames
604 This phase concerns itself with removing directory entries
606 error conditioned inodes
607 from Phase 1 and Phase 1b.
608 This section lists error conditions resulting from
609 root inode mode and status,
610 directory inode pointers in range,
611 and directory entries pointing to bad inodes,
612 and directory integrity checks.
613 All errors in this phase are fatal if the file system is being preen'ed,
614 except for directories not being a multiple of the blocks size
615 and extraneous hard links.
618 .B "ROOT INODE UNALLOCATED (ALLOCATE)"
620 The root inode (usually inode number 2) has no allocate mode bits.
621 This should never happen.
623 Possible responses to the ALLOCATE prompt are:
625 allocate inode 2 as the root inode.
626 The files and directories usually found in the root will be recovered
627 in Phase 3 and put into
629 If the attempt to allocate the root fails,
631 will exit with the message:
633 .B "CANNOT ALLOCATE ROOT INODE" .
639 .B "ROOT INODE NOT DIRECTORY (REALLOCATE)"
641 The root inode (usually inode number 2)
642 is not directory inode type.
644 Possible responses to the REALLOCATE prompt are:
646 clear the existing contents of the root inode
648 The files and directories usually found in the root will be recovered
649 in Phase 3 and put into
651 If the attempt to allocate the root fails,
653 will exit with the message:
655 .B "CANNOT ALLOCATE ROOT INODE" .
658 will then prompt with
661 Possible responses to the FIX prompt are:
663 replace the root inode's type to be a directory.
664 If the root inode's data blocks are not directory blocks,
665 many error conditions will be produced.
667 terminate the program.
670 .B "DUPS/BAD IN ROOT INODE (REALLOCATE)"
672 Phase 1 or Phase 1b have found duplicate blocks
673 or bad blocks in the root inode (usually inode number 2) for the file system.
675 Possible responses to the REALLOCATE prompt are:
677 clear the existing contents of the root inode
679 The files and directories usually found in the root will be recovered
680 in Phase 3 and put into
682 If the attempt to allocate the root fails,
684 will exit with the message:
686 .B "CANNOT ALLOCATE ROOT INODE" .
689 will then prompt with
692 Possible responses to the CONTINUE prompt are:
696 error condition in the root inode and
697 attempt to continue to run the file system check.
698 If the root inode is not correct,
699 then this may result in many other error conditions.
701 terminate the program.
704 .B "NAME TOO LONG \fIF\fP"
706 An excessively long path name has been found.
707 This usually indicates loops in the file system name space.
708 This can occur if the super user has made circular links to directories.
709 The offending links must be removed (by a guru).
712 .B "I OUT OF RANGE I=\fII\fP NAME=\fIF\fP (REMOVE)"
714 A directory entry \fIF\fP has an inode number \fII\fP that is greater than
715 the end of the inode list.
717 Possible responses to the REMOVE prompt are:
719 the directory entry \fIF\fP is removed.
721 ignore this error condition.
724 .B "UNALLOCATED I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP \fItype\fP=\fIF\fP (REMOVE)"
726 A directory or file entry \fIF\fP points to an unallocated inode \fII\fP.
727 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP, modify time \fIT\fP,
728 and name \fIF\fP are printed.
730 Possible responses to the REMOVE prompt are:
732 the directory entry \fIF\fP is removed.
734 ignore this error condition.
737 .B "DUP/BAD I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP \fItype\fP=\fIF\fP (REMOVE)"
739 Phase 1 or Phase 1b have found duplicate blocks or bad blocks
740 associated with directory or file entry \fIF\fP, inode \fII\fP.
741 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP, modify time \fIT\fP,
742 and directory name \fIF\fP are printed.
744 Possible responses to the REMOVE prompt are:
746 the directory entry \fIF\fP is removed.
748 ignore this error condition.
751 .B "ZERO LENGTH DIRECTORY I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (REMOVE)"
753 A directory entry \fIF\fP has a size \fIS\fP that is zero.
754 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP, modify time \fIT\fP,
755 and directory name \fIF\fP are printed.
757 Possible responses to the REMOVE prompt are:
759 the directory entry \fIF\fP is removed;
760 this will always invoke the BAD/DUP error condition in Phase 4.
762 ignore this error condition.
765 .B "DIRECTORY TOO SHORT I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (FIX)"
767 A directory \fIF\fP has been found whose size \fIS\fP
768 is less than the minimum size directory.
769 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP, modify time \fIT\fP,
770 and directory name \fIF\fP are printed.
772 Possible responses to the FIX prompt are:
774 increase the size of the directory to the minimum directory size.
776 ignore this directory.
779 .B "DIRECTORY \fIF\fP LENGTH \fIS\fP NOT MULTIPLE OF \fIB\fP (ADJUST)
781 A directory \fIF\fP has been found with size \fIS\fP that is not
782 a multiple of the directory blocksize \fIB\fP.
784 Possible responses to the ADJUST prompt are:
786 the length is rounded up to the appropriate block size.
787 This error can occur on 4.2BSD file systems.
788 Thus when preen'ing the file system only a warning is printed
789 and the directory is adjusted.
791 ignore the error condition.
794 .B "DIRECTORY CORRUPTED I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (SALVAGE)"
796 A directory with an inconsistent internal state has been found.
798 Possible responses to the FIX prompt are:
800 throw away all entries up to the next directory boundary (usually 512-byte)
802 This drastic action can throw away up to 42 entries,
803 and should be taken only after other recovery efforts have failed.
805 skip up to the next directory boundary and resume reading,
806 but do not modify the directory.
809 .B "BAD INODE NUMBER FOR `.' I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (FIX)"
811 A directory \fII\fP has been found whose inode number for `.' does
812 does not equal \fII\fP.
814 Possible responses to the FIX prompt are:
816 change the inode number for `.' to be equal to \fII\fP.
818 leave the inode number for `.' unchanged.
821 .B "MISSING `.' I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (FIX)"
823 A directory \fII\fP has been found whose first entry is unallocated.
825 Possible responses to the FIX prompt are:
827 build an entry for `.' with inode number equal to \fII\fP.
829 leave the directory unchanged.
832 .B "MISSING `.' I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP"
834 .B "CANNOT FIX, FIRST ENTRY IN DIRECTORY CONTAINS \fIF\fP"
836 A directory \fII\fP has been found whose first entry is \fIF\fP.
838 cannot resolve this problem.
839 The file system should be mounted and the offending entry \fIF\fP
841 The file system should then be unmounted and
846 .B "MISSING `.' I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP"
848 .B "CANNOT FIX, INSUFFICIENT SPACE TO ADD `.'"
850 A directory \fII\fP has been found whose first entry is not `.'.
852 cannot resolve this problem as it should never happen.
856 .B "EXTRA `.' ENTRY I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (FIX)"
858 A directory \fII\fP has been found that has more than one entry for `.'.
860 Possible responses to the FIX prompt are:
862 remove the extra entry for `.'.
864 leave the directory unchanged.
867 .B "BAD INODE NUMBER FOR `..' I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (FIX)"
869 A directory \fII\fP has been found whose inode number for `..' does
870 does not equal the parent of \fII\fP.
872 Possible responses to the FIX prompt are:
874 change the inode number for `..' to be equal to the parent of \fII\fP
875 (``\fB..\fP'' in the root inode points to itself).
877 leave the inode number for `..' unchanged.
880 .B "MISSING `..' I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (FIX)"
882 A directory \fII\fP has been found whose second entry is unallocated.
884 Possible responses to the FIX prompt are:
886 build an entry for `..' with inode number equal to the parent of \fII\fP
887 (``\fB..\fP'' in the root inode points to itself).
889 leave the directory unchanged.
892 .B "MISSING `..' I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP"
894 .B "CANNOT FIX, SECOND ENTRY IN DIRECTORY CONTAINS \fIF\fP"
896 A directory \fII\fP has been found whose second entry is \fIF\fP.
898 cannot resolve this problem.
899 The file system should be mounted and the offending entry \fIF\fP
901 The file system should then be unmounted and
906 .B "MISSING `..' I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP"
908 .B "CANNOT FIX, INSUFFICIENT SPACE TO ADD `..'"
910 A directory \fII\fP has been found whose second entry is not `..'.
912 cannot resolve this problem.
913 The file system should be mounted and the second entry in the directory
915 The file system should then be unmounted and
920 .B "EXTRA `..' ENTRY I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP DIR=\fIF\fP (FIX)"
922 A directory \fII\fP has been found that has more than one entry for `..'.
924 Possible responses to the FIX prompt are:
926 remove the extra entry for `..'.
928 leave the directory unchanged.
931 .B "\fIN\fP IS AN EXTRANEOUS HARD LINK TO A DIRECTORY \fID\fP (REMOVE)
934 has found a hard link, \fIN\fP, to a directory, \fID\fP.
935 When preen'ing the extraneous links are ignored.
937 Possible responses to the REMOVE prompt are:
939 delete the extraneous entry, \fIN\fP.
941 ignore the error condition.
944 .B "BAD INODE \fIS\fP TO DESCEND"
946 An internal error has caused an impossible state \fIS\fP to be passed to the
947 routine that descends the file system directory structure.
953 .B "BAD RETURN STATE \fIS\fP FROM DESCEND"
955 An internal error has caused an impossible state \fIS\fP to be returned
956 from the routine that descends the file system directory structure.
962 .B "BAD STATE \fIS\fP FOR ROOT INODE"
964 An internal error has caused an impossible state \fIS\fP to be assigned
970 Phase 3 \- Check Connectivity
972 This phase concerns itself with the directory connectivity seen in
974 This section lists error conditions resulting from
975 unreferenced directories,
981 .B "UNREF DIR I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP (RECONNECT)"
983 The directory inode \fII\fP was not connected to a directory entry
984 when the file system was traversed.
985 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP, and
986 modify time \fIT\fP of directory inode \fII\fP are printed.
987 When preen'ing, the directory is reconnected if its size is non-zero,
988 otherwise it is cleared.
990 Possible responses to the RECONNECT prompt are:
992 reconnect directory inode \fII\fP to the file system in the
993 directory for lost files (usually \fIlost+found\fP).
996 error condition in Phase 3
997 if there are problems connecting directory inode \fII\fP to \fIlost+found\fP.
998 This may also invoke the CONNECTED error condition in Phase 3 if the link
1001 ignore this error condition.
1002 This will always invoke the UNREF error condition in Phase 4.
1005 .B "NO lost+found DIRECTORY (CREATE)"
1009 directory in the root directory of the file system;
1012 tries to create a \fIlost+found\fP directory.
1014 Possible responses to the CREATE prompt are:
1016 create a \fIlost+found\fP directory in the root of the file system.
1017 This may raise the message:
1019 .B "NO SPACE LEFT IN / (EXPAND)"
1021 See below for the possible responses.
1022 Inability to create a \fIlost+found\fP directory generates the message:
1024 .B "SORRY. CANNOT CREATE lost+found DIRECTORY"
1026 and aborts the attempt to linkup the lost inode.
1027 This will always invoke the UNREF error condition in Phase 4.
1029 abort the attempt to linkup the lost inode.
1030 This will always invoke the UNREF error condition in Phase 4.
1033 .B "lost+found IS NOT A DIRECTORY (REALLOCATE)"
1039 Possible responses to the REALLOCATE prompt are:
1041 allocate a directory inode, and change \fIlost+found\fP to reference it.
1042 The previous inode reference by the \fIlost+found\fP name is not cleared.
1043 Thus it will either be reclaimed as an UNREF'ed inode or have its
1044 link count ADJUST'ed later in this Phase.
1045 Inability to create a \fIlost+found\fP directory generates the message:
1047 .B "SORRY. CANNOT CREATE lost+found DIRECTORY"
1049 and aborts the attempt to linkup the lost inode.
1050 This will always invoke the UNREF error condition in Phase 4.
1052 abort the attempt to linkup the lost inode.
1053 This will always invoke the UNREF error condition in Phase 4.
1056 .B "NO SPACE LEFT IN /lost+found (EXPAND)"
1058 There is no space to add another entry to the
1060 directory in the root directory
1064 directory is expanded.
1066 Possible responses to the EXPAND prompt are:
1070 directory is expanded to make room for the new entry.
1071 If the attempted expansion fails
1075 .B "SORRY. NO SPACE IN lost+found DIRECTORY"
1077 and aborts the attempt to linkup the lost inode.
1078 This will always invoke the UNREF error condition in Phase 4.
1079 Clean out unnecessary entries in
1081 This error is fatal if the file system is being preen'ed.
1083 abort the attempt to linkup the lost inode.
1084 This will always invoke the UNREF error condition in Phase 4.
1087 .B "DIR I=\fII1\fP CONNECTED. PARENT WAS I=\fII2\fP"
1089 This is an advisory message indicating a directory inode \fII1\fP was
1090 successfully connected to the
1093 The parent inode \fII2\fP of the directory inode \fII1\fP is
1094 replaced by the inode number of the
1099 .B "DIRECTORY \fIF\fP LENGTH \fIS\fP NOT MULTIPLE OF \fIB\fP (ADJUST)
1101 A directory \fIF\fP has been found with size \fIS\fP that is not
1102 a multiple of the directory blocksize \fIB\fP
1103 (this can reoccur in Phase 3 if it is not adjusted in Phase 2).
1105 Possible responses to the ADJUST prompt are:
1107 the length is rounded up to the appropriate block size.
1108 This error can occur on 4.2BSD file systems.
1109 Thus when preen'ing the file system only a warning is printed
1110 and the directory is adjusted.
1112 ignore the error condition.
1115 .B "BAD INODE \fIS\fP TO DESCEND"
1117 An internal error has caused an impossible state \fIS\fP to be passed to the
1118 routine that descends the file system directory structure.
1123 Phase 4 \- Check Reference Counts
1125 This phase concerns itself with the link count information
1126 seen in Phase 2 and Phase 3.
1127 This section lists error conditions resulting from
1132 incorrect link counts for files, directories, symbolic links, or special files,
1133 unreferenced files, symbolic links, and directories,
1134 and bad or duplicate blocks in files, symbolic links, and directories.
1135 All errors in this phase are correctable if the file system is being preen'ed
1136 except running out of space in the \fIlost+found\fP directory.
1139 .B "UNREF FILE I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP (RECONNECT)"
1141 Inode \fII\fP was not connected to a directory entry
1142 when the file system was traversed.
1143 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP, and
1144 modify time \fIT\fP of inode \fII\fP are printed.
1145 When preen'ing the file is cleared if either its size or its
1147 otherwise it is reconnected.
1149 Possible responses to the RECONNECT prompt are:
1151 reconnect inode \fII\fP to the file system in the directory for
1152 lost files (usually \fIlost+found\fP).
1155 error condition in Phase 4
1156 if there are problems connecting inode \fII\fP to
1159 ignore this error condition.
1160 This will always invoke the CLEAR error condition in Phase 4.
1165 The inode mentioned in the immediately previous error condition can not be
1167 This cannot occur if the file system is being preen'ed,
1168 since lack of space to reconnect files is a fatal error.
1170 Possible responses to the CLEAR prompt are:
1172 de-allocate the inode mentioned in the immediately previous error condition by zeroing its contents.
1174 ignore this error condition.
1177 .B "NO lost+found DIRECTORY (CREATE)"
1181 directory in the root directory of the file system;
1184 tries to create a \fIlost+found\fP directory.
1186 Possible responses to the CREATE prompt are:
1188 create a \fIlost+found\fP directory in the root of the file system.
1189 This may raise the message:
1191 .B "NO SPACE LEFT IN / (EXPAND)"
1193 See below for the possible responses.
1194 Inability to create a \fIlost+found\fP directory generates the message:
1196 .B "SORRY. CANNOT CREATE lost+found DIRECTORY"
1198 and aborts the attempt to linkup the lost inode.
1199 This will always invoke the UNREF error condition in Phase 4.
1201 abort the attempt to linkup the lost inode.
1202 This will always invoke the UNREF error condition in Phase 4.
1205 .B "lost+found IS NOT A DIRECTORY (REALLOCATE)"
1211 Possible responses to the REALLOCATE prompt are:
1213 allocate a directory inode, and change \fIlost+found\fP to reference it.
1214 The previous inode reference by the \fIlost+found\fP name is not cleared.
1215 Thus it will either be reclaimed as an UNREF'ed inode or have its
1216 link count ADJUST'ed later in this Phase.
1217 Inability to create a \fIlost+found\fP directory generates the message:
1219 .B "SORRY. CANNOT CREATE lost+found DIRECTORY"
1221 and aborts the attempt to linkup the lost inode.
1222 This will always invoke the UNREF error condition in Phase 4.
1224 abort the attempt to linkup the lost inode.
1225 This will always invoke the UNREF error condition in Phase 4.
1228 .B "NO SPACE LEFT IN /lost+found (EXPAND)"
1230 There is no space to add another entry to the
1232 directory in the root directory
1236 directory is expanded.
1238 Possible responses to the EXPAND prompt are:
1242 directory is expanded to make room for the new entry.
1243 If the attempted expansion fails
1247 .B "SORRY. NO SPACE IN lost+found DIRECTORY"
1249 and aborts the attempt to linkup the lost inode.
1250 This will always invoke the UNREF error condition in Phase 4.
1251 Clean out unnecessary entries in
1253 This error is fatal if the file system is being preen'ed.
1255 abort the attempt to linkup the lost inode.
1256 This will always invoke the UNREF error condition in Phase 4.
1259 .B "LINK COUNT \fItype\fP I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP COUNT=\fIX\fP SHOULD BE \fIY\fP (ADJUST)"
1261 The link count for inode \fII\fP,
1262 is \fIX\fP but should be \fIY\fP.
1263 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP, and modify time \fIT\fP
1265 When preen'ing the link count is adjusted unless the number of references
1266 is increasing, a condition that should never occur unless precipitated
1267 by a hardware failure.
1268 When the number of references is increasing under preen mode,
1270 exits with the message:
1272 .B "LINK COUNT INCREASING"
1274 Possible responses to the ADJUST prompt are:
1276 replace the link count of file inode \fII\fP with \fIY\fP.
1278 ignore this error condition.
1281 .B "UNREF \fItype\fP I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP (CLEAR)"
1283 Inode \fII\fP, was not connected to a directory entry when the
1284 file system was traversed.
1285 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP,
1286 and modify time \fIT\fP of inode \fII\fP
1289 this is a file that was not connected because its size or link count was zero,
1290 hence it is cleared.
1292 Possible responses to the CLEAR prompt are:
1294 de-allocate inode \fII\fP by zeroing its contents.
1296 ignore this error condition.
1299 .B "BAD/DUP \fItype\fP I=\fII\fP OWNER=\fIO\fP MODE=\fIM\fP SIZE=\fIS\fP MTIME=\fIT\fP (CLEAR)"
1301 Phase 1 or Phase 1b have found duplicate blocks
1302 or bad blocks associated with
1304 The owner \fIO\fP, mode \fIM\fP, size \fIS\fP,
1305 and modify time \fIT\fP of inode \fII\fP
1307 This error cannot arise when the file system is being preen'ed,
1308 as it would have caused a fatal error earlier.
1310 Possible responses to the CLEAR prompt are:
1312 de-allocate inode \fII\fP by zeroing its contents.
1314 ignore this error condition.
1316 Phase 5 - Check Cyl groups
1318 This phase concerns itself with the free-block and used-inode maps.
1319 This section lists error conditions resulting from
1320 allocated blocks in the free-block maps,
1321 free blocks missing from free-block maps,
1322 and the total free-block count incorrect.
1323 It also lists error conditions resulting from
1324 free inodes in the used-inode maps,
1325 allocated inodes missing from used-inode maps,
1326 and the total used-inode count incorrect.
1329 .B "CG \fIC\fP: BAD MAGIC NUMBER"
1331 The magic number of cylinder group \fIC\fP is wrong.
1332 This usually indicates that the cylinder group maps have been destroyed.
1333 When running manually the cylinder group is marked as needing
1334 to be reconstructed.
1335 This error is fatal if the file system is being preen'ed.
1338 .B "BLK(S) MISSING IN BIT MAPS (SALVAGE)"
1340 A cylinder group block map is missing some free blocks.
1341 During preen'ing the maps are reconstructed.
1343 Possible responses to the SALVAGE prompt are:
1345 reconstruct the free block map.
1347 ignore this error condition.
1350 .B "SUMMARY INFORMATION BAD (SALVAGE)"
1352 The summary information was found to be incorrect.
1354 the summary information is recomputed.
1356 Possible responses to the SALVAGE prompt are:
1358 reconstruct the summary information.
1360 ignore this error condition.
1363 .B "FREE BLK COUNT(S) WRONG IN SUPERBLOCK (SALVAGE)"
1365 The superblock free block information was found to be incorrect.
1367 the superblock free block information is recomputed.
1369 Possible responses to the SALVAGE prompt are:
1371 reconstruct the superblock free block information.
1373 ignore this error condition.
1377 Once a file system has been checked, a few cleanup functions are performed.
1378 This section lists advisory messages about
1380 and modify status of the file system.
1383 .B "\fIV\fP files, \fIW\fP used, \fIX\fP free (\fIY\fP frags, \fIZ\fP blocks)"
1385 This is an advisory message indicating that
1386 the file system checked contained
1388 \fIW\fP fragment sized blocks leaving
1389 \fIX\fP fragment sized blocks free in the file system.
1390 The numbers in parenthesis breaks the free count down into
1391 \fIY\fP free fragments and
1392 \fIZ\fP free full sized blocks.
1395 .B "***** REBOOT UNIX *****"
1397 This is an advisory message indicating that
1398 the root file system has been modified by
1400 If UNIX is not rebooted immediately,
1403 may be undone by the in-core copies of tables
1407 will exit with a code of 4.
1408 The standard auto-reboot script distributed with 4.3BSD
1409 interprets an exit code of 4 by issuing a reboot system call.
1412 .B "***** FILE SYSTEM WAS MODIFIED *****"
1414 This is an advisory message indicating that
1415 the current file system was modified by
1417 If this file system is mounted or is the current root file system,
1419 should be halted and UNIX rebooted.
1420 If UNIX is not rebooted immediately,
1423 may be undone by the in-core copies of tables