1 Mon Oct 18 14:56:22 1999 H.J. Lu <hjl@lucon.org>
3 * Initial version 0.1 released.
5 * configure.in (VERSION): Set to "nfs-utils 0.1".
6 * configure: Regenerated.
8 Mon Oct 18 14:54:57 1999 H.J. Lu <hjl@lucon.org>
10 * utils/mountd/mountd.c (get_exportlist): Cleanup.
12 * utils/exportfs/exportfs.c (unexport_all): Unexport from
13 kernel only if the entry is exported to kernel.
14 (unexportfs): Likewise.
16 Wed Sep 08 16:49:32 1999 Neil Brown <neilb@cse.unsw.edu.au>
18 1/ utils/mountd/rmtab.c::mountlist_list
20 This routine stats the rmtab file to see if it has changed. It
21 if has, it cleans up it's old copy of the data. But it still
22 always re-read the file, thus returning multiple copies of the
23 data on consecutive calls without intervening changes.
24 "Showmount -a" didn't show this as it appears to sort/unique the
25 data, but 'strace showmount -a' showed that the size of the
26 datagram that it received grew.
28 I moved the getrmtabent loop inside the mtime test.
30 2/ utils/exportfs/exportfs.c
32 Many routines used the m_path field of m_export instead of
34 According to the comment in nfslib.h, m_path should only
35 be used when processing a mount request (i.e. in mountd)
36 where the mountpoint may be a subdirectory of the export point.
38 I changed all occurances of m_path to e_path
41 3/ utils/exportfs/exportfs.c:main
43 extra arguments are not meaningful with -a or -r, but
44 exportfs accepted them and then ignored the -a/-r, expect that
45 -r would still unexport everything first.
47 I generate an error if there are extra args and f_all
49 4/ utils/exportfs/exportfs.c:main
50 extract dump out as a special case.
52 5/ utils/exportfs/exportfs.c
53 made f_reexport a local variable.
56 6/ utils/exportfs/exportfs.c:main,exportall
58 support/export/rmtab.c
59 only mayexport on newly created entries, don't set xtabent at all
61 7/ support/include/nfslib.h
63 add #define _PATH_PROC_EXPORTS to be /proc/fs/nds/exports
65 8/ support/export/xtab.c
67 xtab_mount_read loads data from _PATH_PROC_EXPORTS if it exists,
71 9/ support/export/xtab.c
73 xtab_mount_read now sets m_exported, and NOT
76 removed the append arguement from xtab_write as it was
79 added is_export flag to xtab_write similar to xtab_read
80 if is_export, only write entries with m_xtabent or m_addxtab
81 if !is_export, only write entries with m_exported
83 10/ support/export/export.c::export_allowed_internal
85 added test for exp->m_mayexport, as the export tree
86 may have entries that are no longer allowed to be exported,
87 and so shouldn't caused deduced exported by rmtab_read
89 11/ utils/exportfs/exportfs.c::main
90 error checking of flags.
92 12/ utils/exportfs/exportfs.c
94 total rewrite of export and unexport logic.
96 - build an exportslist of valid exports, based on
97 current etab file and arguments,
98 - read rmtab to instantiate relevant wild card entries
99 - read etab to find out what is currently exported
100 - synchronise intention with reality
101 - write out etab and xtab
104 discard the m_addxtab flag
105 add m_changed flag so we know what to report in exportfs
107 14/ utils/mountd/auth.c:auth_authenticate
109 the value returned by gethostbyaddr was trusted.
111 It now follows this with a call to gethostbyname
112 and checks that the address is in the list.
114 15/ support/export/nfsctl.c::cltsetup,expsetup
116 force client names to lowercase as kernel is
119 16/ quietened a few compiler warnings
121 17/ support/export/client:client_lookup
123 look for pre-existing client with same name before creating
126 18/ support/include/exportfs.h
128 The ordering of the MCL_* enum was:
129 ANONYMOUS, FQDN, SUBNETWORK, WILDCARD, NETGROUP
131 I moved ANONYMOUS to the end.
133 The ordering is significant when an export entry is being searched for to
134 match a given address. There are two problems with ANONYMOUS being first.
136 1/ if a directory is exported rw to a couple of hosts and ro to everyone else,
137 then the ro case will always be found first and the privileged hosts won't get
139 2/ When mountd gets a request to mount an ANONYMOUSly exported tree, it creates a FQDN
140 export entry for the specific host, and writes it to xtab.
141 When another request comes from the same host, the ANONYMOUS entry is found again,
142 before the new FQDN entry, so it creates another FQDN entry and writes it to xtab
143 again. If causes bloat in xtab.
145 Putting ANONYMOUS at the end reflects it's nature as a catch-all
147 19/ utils/exportfs/exportfs.man
148 many updates to the man page to reflect changes to the code
150 -----------------------
156 - allow exportfs to modify rmtab file
157 - make sure kernel never gets two clients with same IP address
158 - possible kernel should reject
159 - needs to be some way to lookup client in kernel by IP address
160 - maybe get kernel to do case-insensitive comparisons on client names
161 - remove unused clients from kernel
163 - change etab to xtab and xtab to xtab.active
165 - timestamp and/or statd-stamp in rmtab for removing old entries.
167 Mon Oct 18 11:48:07 1999 H.J. Lu <hjl@lucon.org>
169 * linux-nfs: New directory.
170 * linux-nfs/ChangeLog: Moved from ..
171 * linux-nfs/INSTALL: Likewise.
172 * linux-nfs/KNOWNBUGS: Likewise.
173 * linux-nfs/NEW: Likewise.
174 * linux-nfs/README: Likewise.
175 * linux-nfs/THANKS: Likewise.
176 * linux-nfs/TODO: Likewise.
178 * Starting from knfsd 1.4.7.