diff options
Diffstat (limited to 'ChangeLog')
-rw-r--r-- | ChangeLog | 1118 |
1 files changed, 1089 insertions, 29 deletions
diff --git a/ChangeLog b/ChangeLog index 12f2c5fb8e65..d996f8e75082 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,14 +1,564 @@ -2006-05-11 Erez Zadok <ezk@cs.sunysb.edu> +2014-10-28 Christos Zoulas <christos@zoulas.com> ******************************************************************* - *** Released am-utils-6.1.5 *** + *** Released am-utils-6.2 *** ******************************************************************* -2006-05-10 Erez Zadok <ezk@cs.sunysb.edu> + * add ylwrap.amd, sick and tired of patching broken ylwrap. + +2014-09-05 Christos Zoulas <christos@zoulas.com> + + * factor out the size check for both readdir methods + (raven at themaw.net) + +2014-09-03 Christos Zoulas <christos@zoulas.com> + + * configure test for 64 bit xdr function (raven at themaw.net) + +2014-09-01 Christos Zoulas <christos@zoulas.com> + + * bug fix from raven at themaw.net to prevent writing v3 handle + to a v2 structure. + +2014-07-21 Christos Zoulas <christos@zoulas.com> + + * fix permission problems with access + * auto_nfs_version = 3 now works for me. + +2014-07-20 Christos Zoulas <christos@zoulas.com> + + * refresh some autoconf files + * force autoconf version since earlier versions don't work + * changes from raven at themaw.net to support NFSv3 for automounted + toplevel filesystems presented to the list in May. A few bug fixes + and integration changes by me. The code now works well enough to + mount the filesystem with NFSv3 and automount children but readdir + does not work yet (so no listing of automounted filesystems). + The new code is not enabled by default. To enable use + auto_nfs_version = 3 + in your configuration file. + +2014-03-26 Christos Zoulas <christos@zoulas.com> + + From Ian Kent <ikent@redhat.com> + + * use mktemp in scripts/fixrmtab + * fix documentatio nits + * [linux] Add ext2, ext3, ext4 specific mount options + * fix tcpwrappers (hosts.{allow,deny} problems for specific host + entries + +2014-03-20 Christos Zoulas <christos@zoulas.com> + + + * don't background autofs umount + * avoid fh NULL dereference on umount + * handle ENOENT umount return for autofs mounts + * fix error message in get_nfs_version + * fix debug log deadlock + * [linux] retry umount on EBUSY + * handle failed umount on exit + +2014-03-09 Christos Zoulas <christos@zoulas.com> + + * Remove BSD advertising clause. + * Remove expn program which is not used by anything + +2013-12-05 Christos Zoulas <christos@zoulas.com> + + * Change the way we handle multiple lex scanners and yacc parsers + in a single program. Our old patch to ylwrap does not work anymore + because the bison parsers need shared symbols that start with yy + and we can't easily select which ones work. So now we use -P and + -p to let lex and yacc do the work for us. This requires a patch + to ylwrap, but it is smaller. + + +2013-12-05 Christos Zoulas <christos@zoulas.com> + + * Change the way we handle multiple lex scanners and yacc parsers + in a single program. Our old patch to ylwrap does not work anymore + because the bison parsers need shared symbols that start with yy + and we can't easily select which ones work. So now we use -P and + -p to let lex and yacc do the work for us. This requires a patch + to ylwrap, but it is smaller. + +2013-05-14 Christos Zoulas <christos@zoulas.com> + + * update to handle new autoconf and regen files + +2012-11-23 Christos Zoulas <christos@zoulas.com> + + * recognize mountport=, noacl options + * simplify the bootstrap process, by forcing + all the files to be rebuild with autoreconf -f - i + +2012-10-03 Christos Zoulas <christos@zoulas.com> + + * add missing ops_nfs4.c to the Makefile.am + +2012-09-14 Christos Zoulas <christos@zoulas.com> + + * add locking option handling for lustre + * fix noatime, nodiratime, mandlock for linux + +2012-07-25 Christos Zoulas <christos@zoulas.com> + + * bug fixes for lustre rhost parsing + +2012-06-27 Christos Zoulas <christos@zoulas.com> + + * Disable 1/2-baked sun map support to avoid SIGSEGV + +2012-06-16 Christos Zoulas <christos@zoulas.com> + + ******************************************************************* + *** Released am-utils-6.2-rc1 *** + ******************************************************************* + + * Fix libtool.m4 to work with later versions of autoconf. + * In buildall, use eval so that quoted strings work + * Centralize nfs version checking function, and introduce + new configuration variable nfs_vers_ping, which is the + max version we ping for. This is not documented yet, because + there might be a better way separate the max version we can + use vs the max version we ping for. + +2012-06-09 Christos Zoulas <christos@zoulas.com> + + * fix issue with parsing nfsv4 security option strings + +2012-05-01 Christos Zoulas <christos@zoulas.com> + + * Handle EPERM return from v4 file handles + in old linux kernels + +2012-03-09 Christos Zoulas <christos@zoulas.com> + + * fix linux autofs detection + * fix wrong types in new amq mount xdr processing + +2012-01-25 Christos Zoulas <christos@zoulas.com> + + * FreeBSD portability fixes + +2012-01-23 Christos Zoulas <christos@zoulas.com> + + * Avoid type punned warnings + +2011-12-29 Christos Zoulas <christos@zoulas.com> + + * fix bug I introduced in get_nfs_version that caused it to return 0 + * add direct support for ext{2,3,4} + +2011-12-15 Christos Zoulas <christos@zoulas.com> + + * Include <linux/fs.h> on linux, to grab extra mount flags + * Add support for lustre + +2011-11-28 Christos Zoulas <christos@zoulas.com> + + * For nfsl mounts, match against either the short local + host name or the fqdn name. + +2011-10-06 Christos Zoulas <christos@zoulas.com> + + * Add NFSv4 support (loosely based on Ramya's code). + Currently if the NFSv4 mount fails with ENOENT we fail + back to NFSv3, unless -DNO_FALLBACK. + * Replace strdup with xstrdup. It is not nice to replace + system functions with ones that behave differently. + * Add various consistency checks (caller function checks + variable for NULL, but callee does not). + * Fix error handling in various paths. + * Don't check before using XFREE(). free(NULL) works. + +2011-09-15 Christos Zoulas <christos@zoulas.com> + + * amq -i addition (prints info about the map files) + +2011-09-14 Christos Zoulas <christos@zoulas.com> + + * Solaris OS and compiler fixes from: + Hofmann, Frank <Frank.Hofmann@morganstanley.com> + Sayle, Alex <Alex.Sayle@morganstanley.com> + - (void)&var generates warnings, removed + - __inline is invalid, changed with inline + - <sys/socket.h> and other headers have extra dependendcies + ` +2011-08-11 Christos Zoulas <christos@zoulas.com> + + * fix potential buffer overflow in expand_error() + (Richard Kettlewell). Does not seem to be exploitable. + + * simplify %option parsing to appease ancient flex. + +2011-06-21 Christos Zoulas <christos@zoulas.com> + + * add support for linux autofs version 5 (untested) + +2011-06-18 Christos Zoulas <christos@zoulas.com> + + * roll back to an older version of libtool that works + + * newer versions of config.guess and config.sub + + * minor fixes to buildall + +2011-05-19 Christos Zoulas <christos@zoulas.com> + + * m4/macros/libtool.m4: for AIX AC_LANG_PROGRAM is incorrectly + used and the latest autoconf gives an error. For now just + quote it. + + * configure.in: Fedora Core 15 and RedHat 6 now don't define + af_family_t in <linux/socket.h>, but in <bits/sockaddr.h>. + Include <sys/socket.h> so that we get <bits/sockaddr.h> + indirectly. This caused the <linux/nfs_mount.h> detection + test to fail. + +2010-11-24 Erez Zadok <ezk@fsl.cs.sunysb.edu> + + * m4/macros/os_cflags.m4: for linux, define _GNU_SOURCE (to get + definitions for O_LARGEFILE and more). + +2010-11-20 Christos Zoulas <christos@zoulas.com> + + * Update news items, add release readme, mark as rc1 + + * Fix addopts code to work with options that are key=value + when the built-in hasmntopt() code is not used. The problem + was that the code passed "key=" to hasmntopt not "key". + +2009-12-10 Christos Zoulas <christos@zoulas.com> + + * Copy all fields of the am_opts structure + When we introduced am_loc we also added a function to + create a duplicate of an am_opts structure. As it turns + out, that only made a partial copy of the struct and some + fields (e.g. opt_cache) was left blank in the new structure. + Unfortunately any code that was checking the options + through the mntfs structure was mislead by the partial + copy: caching was disabled completely, for example, and + since browsing relies on caching being enabled it wasn't + working either. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * Initialize RPC authentication flavor to AUTH_UNIX instead + of zero. RHEL 5 uses version 6 of the mount_arg structure + that includes the desired RPC authorization flavour. + Unfortunately amd initializes that value to zero, meaning + AUTH_NULL, resulting in all NFS servers denying operations. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * Make sure to remove nodes in the proper order when going + down. Depending on what order the nodes got created it's + possible that the parent of a node has a bigger am_mapno + (index in exported_ap[]) so that it gets freed before + its child while the child's am_parent pointer is still + pointing to the already freed block of memory. + This change makes sure that umount_exported() cleans up + all children of a node first before freeing the node. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * Fix Linux-specific stuff in ops_tmpfs.c + AUTOFS_TMPFS_FS_FLAGS is defined only in the Linux-specific + conf_linux.c file, so it cannot be built on Solaris. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * Add missing parenthesis to fix defined(XXX + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * In the previous commit we've already made a change that + makes this code unnecessary since the node is returned + even if the backing file server is known to be down. + Since the code path is unused it should be removed. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * amfs_generic_lookup_child() tries to find an alternate + file server if it detects that the current one is down. + However, in the comparison used to decide if a possible + mount is the same as the current one it compares the am_loc + pointers instead of the actual mntfs pointers. Since + amfs_lookup_one_location() creates a brand new am_loc + structure for the returned mntfs structure the am_loc + pointer will be different even if the mntfs it's pointing + to is the same. + This tricks amfs_generic_lookup_child() think it has + found an alternate file server while it hasn't and thus + starts multiple racing mounts for the very same node and + mntfs. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * In the new am_loc scheme mounted mntfs structures still + have their mf_fo structure left intact. So when restart + creates a fake mntfs for an already mounted filesystem it + should not free mf_fo right away. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * Setting the error flag of the mntfs for which the share + is already mounted results in the mount point not being + removed when it times out. (Amd considers it an error + mntfs and thus umount becomse a no-op.) + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * Right now AMD sets all timeout values to their defaults + even if you explicitly set them to zero. This patch + introduces separate flags that store if the respective + values have been set and the code path setting default + values checks these flags instead of testing the values + for zero. [Linux only] + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * Fix mntfs related lookup/(un)mount race conditions, by + introducing a new structure linking an mntfs to a node. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * We have to free the buffers allocated by the XDR routines when + decoding the mount response after copying them to their final place. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + +2009-12-09 Christos Zoulas <christos@zoulas.com> + + * Free continuation in amfs_retry() if we don't call amfs_bgmount() + Normally it's amfs_bgmount() which frees the continuation. However, if + the mount is already in progress and we're in amfs_retry() for another + node we don't try mounting the filesystem once again. Still, we have + to free the continuation as we won't get called again and thus would + leak the continuation structure and our am_loc references. + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + + * Fix use-after-free and double free of mntfs private data + ops_nfs.c(nfs_init) does not clear private data pointer after free(). + This leads to use-after-free and double free in case mount() fails. + (found by valgrind) + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + +2009-11-23 Christos Zoulas <christos@zoulas.com> + + * The temporary filename buffer was leaked in lock_mtab(). + From: Krisztian Kovacs <Kris.Kovacs@morganstanley.com> + +2009-10-27 Christos Zoulas <christos@zoulas.com> + + * Deal with errno.h->sys/errno.h better + +2009-10-26 Christos Zoulas <christos@zoulas.com> + + * Fix issue with autoconf 2.63 + + * Fix compile warnings for gcc-4.4.x + +2009-04-15 Christos Zoulas <christos@zoulas.com> + + * Share the LDAP connection between different maps to save resources + on the LDAP server. (Olaf Flebbe <o.flebbe@science-computing.de>) + +2009-03-20 Christos Zoulas <christos@zoulas.com> + + * Fix largefile re-definitions + + * Fix compilation error caused in previous patch + +2009-02-19 Christos Zoulas <christos@zoulas.com> + + * Add a synchronous unmount amq rpc that will wait for the remote + filesystem to be unmounted, or return an error. Enabled by amq -uu + +2009-01-11 Erez Zadok <ezk@fsl.cs.sunysb.edu> + + * amd/ops_udf.c: don't define functions/variables which may not be + used. + + * amd/ops_tmpfs.c (mount_tmpfs): force gcc not to complain about + unused variables. + + * config.guess.long: properly support Fedora Core. + +2009-01-09 Christos Zoulas <christos@zoulas.com> + + * Fix nfs mounts on linux 2.6.26 by explicitly initializing context. + + * Treat an empty sublink as an unset sublink; this allows us to + say sublink:=; in a spec to override the default. (This produces + Bad selector error, but it works. XXX: We need to decide how to + unset a selector. I proposed an unset keyword a while ago). + +2009-01-06 Christos Zoulas <christos@zoulas.com> + + * re-renerate all generated files with newer versions of the + auto-tools. + + * move INSTALL to INSTALL.am-utils to avoid conflicts + + * fix incompatibilities with new m4 and new autoconf. + + * add UDF filesystem support. + + * fix NetBSD nfs file handle detection. + +2009-01-02 Christos Zoulas <christos@zoulas.com> + + * add nocasetrans, nojoliet, rrcaseins for cd9660 filesystems + +2008-12-03 Christos Zoulas <christos@zoulas.com> + + * allow ldap queries try a comma-separated list of servers given as: + + ldap_hostports = host1:port1,host2,host3:port3 + + original patch from Florian Geyer + +2008-09-19 Christos Zoulas <christos@zoulas.com> + + * the auth_create gid on NetBSD is int + + * deal with lossage from flex -P + + * fix missing/incorrect prototypes + +2008-09-17 Christos Zoulas <christos@zoulas.com> + + * fix manual page section issues (sorting, non-existant) + + * add missing forward declarations + + * avoid null pointer derefs (coverity) + + * more portable endian detection support + + * implement tmpfs support + + * use sizeof var instead of constant + + * fix yyerror declaration and definition + + * fix yy function renaming, and correct some renaming issues. + + * work around automake < 1.9 issue where ylwrap is not being run + for a single yacc or lex file. + + * NetBSD + - handle headers that need other headers to compile + - add exfs and tmpfs specific glue + - add options log, noatime, nocoredump, nodevmtime, softdep, + symperm, union + - NetBSD's new mount system call needs the size of the opaque + filesystem specific structure, but passing 0 for now works. + We would need to change a lot more code to pass the size from + the fs-specific calls, so for now punt. + - recvfrom takes socklen_t * on NetBSD + - handle newer NetBSD using statvfs + - NetBSD calls its ufs ffs + + * Linux + - Most recent kernels have trouble with trying to parse the pid + hostname and do locking. What before was a silent failure, now + there is a warning printed and the mount is rejected. This is + the simplest fix: + (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=479884) + +2007-04-13 Erez Zadok <ezk@cs.sunysb.edu> + + * amd/sun_map.c (sun_append_str): When translating '&' from Sun + automounter maps, replace it with "${key}" in the Amd map, not + with the value of the key at the time. This way '*' maps work + too. + + * config.guess.long: don't use -q option to grep because some + systems don't support it. Bug fix for solaris. + +2007-03-11 Christos Zoulas <christos@zoulas.com> + + * amd/opts.c (expand_op): Don't call xstrlcpy() to truncate a + string. It causes spurious xstrlcpy() syslog() errors. Use + memcpy() and explicitly terminate the string. + +2007-01-20 Erez Zadok <ezk@cs.sunysb.edu> + + * updated copyright year to 2007 on all files. + +2006-12-27 Erez Zadok <ezk@cs.sunysb.edu> + + * minor new port: ia64-hp-hpux11.23 + + * m4/macros/check_libwrap_severity.m4 (LIBS): also test if caller + needs to define rfc931_timeout. + + * amd/sun2amd.c: define rfc931_timeout and allow/deny_severity for + libwrap. + + * amd/amq_svc.c: some versions of libwrap require the caller to + define rfc931_timeout in addition to allow/deny_severity. + + * amd/amfs_toplvl.c (amfs_toplvl_init): properly detect when + forced unmount functionality exists for all systems. + + * libamu/mount_fs.c (compute_nfs_args): reset nfs_args_t->pathconf + only if field exists. + + * conf/nfs_prot/nfs_prot_hpux11.h: prevent HPUX 11.23 from + including <tiuser> or <sys/tiuser> (TLI definitions are already + included from elsewhere). + + * configure.in: detect nfs_args_t->pathconf field. + +2006-11-27 Erez Zadok <ezk@cs.sunysb.edu> + + * doc/Makefile.am: properly publish html files using newer + text2html. + +2006-11-27 Erez Zadok <ezk@shekel.local> + + * doc/Makefile.am: obfuscate emails in ChangeLog before it is + published on web site. + + * obfuscate many more emails in the distro, esp. of files which get + posted on the am-utils web site. + +2006-11-26 Erez Zadok <ezk@shekel.local> + + * minor new port: i386-apple-darwin8.8.1. + + * doc/am-utils.texi: remove obvious URLs used by spammers, and + replace them with links to the am-utils web site, which uses PNG + images to obfuscate the actual email/mailman URLs. + +2006-10-30 Erez Zadok <ezk@cs.sunysb.edu> + + * doc/am-utils.texi (Password maps): minor typo reported by + Christoph P. Kukulies. + +2006-10-26 Erez Zadok <ezk@cs.sunysb.edu> + + * minor new port: i386-pc-linux-fc6. + + * m4/macros/linux_headers.m4: handle newer Linux kernels which + removed UTS_RELEASE from <linux/version.h>. + +2006-10-25 Erez Zadok <ezk@cs.sunysb.edu> + + * minor new ports: i386-pc-linux-centos4.4 and + powerpc-apple-darwin8.7.0. + + * config.guess.long: recognize CentOS Linux. + +2006-06-20 Erez Zadok <ezk@cs.sunysb.edu> + + * minor new port: i386-pc-solaris2.11-nexentaos (GNU/OpenSolaris). + + * config.guess.long: recognize NexentaOS (GNU/OpenSolaris) - * Minor new ports: i386-unknown-freebsd6.1 (RELEASE) + * config.guess, config.sub: updates from latest official GNU + distributions. + +2006-05-11 Erez Zadok <ezk@cs.sunysb.edu> + + * Minor new ports: i386-unknown-freebsd6.1 (RELEASE), powerpc-apple-darwin8.6.0, and i386-unknown-openbsd3.9. +2006-05-10 Erez Zadok <ezk@cs.sunysb.edu> + * libamu/mount_fs.c (compute_nfs_attrcache_flags): fix signedness problems on IRIX-6.5. @@ -188,11 +738,14 @@ same $ac_cv_* variable name). Some systems support one header or another. Patch inspired by work Debian did. -2006-02-10 Erez Zadok <ezk@cs.sunysb.edu> +2006-02-16 Daniel P. Ottavio <ottavio@vm.fsl.cs.sunysb.edu> - ******************************************************************* - *** Released am-utils-6.1.4 *** - ******************************************************************* + * amd/sum_map_tok.l: Fixed a bug in a regular expression that + prevented dashes in hostnames. This was posted in BZ by Thomas + A. Fine. + + * README.sun2amd: Fixed a typo. Sentence mentions use of autofs + type when it should be host type. 2006-01-04 Erez Zadok <ezk@cs.sunysb.edu> @@ -302,9 +855,31 @@ 2005-10-10 Erez Zadok <ezk@cs.sunysb.edu> ******************************************************************* - *** Released am-utils-6.1.3 *** + *** Released am-utils-6.1a1 *** ******************************************************************* + * README.sun2amd: revised. + +2005-10-09 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/info_nis.c (nis_search): Bug fix for bugzilla #428; make + sure that sun_entry2amd() is not called when yp_match() returns an + error, this is done by checking that 'res' is 0. Bug fix (same + BZ#); free the output allocated by yp_match() instead of freeing + the return value of sun_entry2amd(). + + * amd/sun_map.c (sun_entry2amd): Fix a comment. + + * README.sun2amd: Made a first pass at describing the status of + the sun2amd feature. + Moved the date. + +2005-10-09 Erez Zadok <ezk@cs.sunysb.edu> + + * README.sun2amd: place holder. + + * Makefile.am (EXTRA_DIST): distribute new README.sun2amd file. + 2005-10-07 Erez Zadok <ezk@cs.sunysb.edu> * m4/macros/check_mnttab_type.m4: move the test for MOUNT_* to the @@ -343,6 +918,10 @@ * amd/info_exec.c (exec_map_open): break assignment from inside conditional, to work around an IRIX-6.5 cc bug. + * amd/Makefile.am (amd_LDADD, sun2amd_LDADD): put libaminfo.a + before libamu.la. Some linkers (e.g., IRIX-6.5) incorrectly + complain about undefined symbols. + * fsinfo/fsi_util.c (set_ether_if), amd/map.c (unmount_mp), libamu/xutil.c (expand_error), libamu/strutil.c (xsnprintf): avoid comparison between signed and unsigned integers. @@ -479,15 +1058,10 @@ 2005-10-02 Erez Zadok <ezk@cs.sunysb.edu> - * include/am_utils.h: extern for xvsnprintf(). - - * amd/autil.c: move strsplit() to libamu/strutil.c. - * libamu/Makefile.am: use strutil.c, not util.c. * libamu/strutil.c: rename from util.c to explain better purpose - of file. Move xvsnprintf and xsnprintf from xutil.c to this - file. Sync up with 6.2 version of file. + of file. Move xvsnprintf and xsnprintf from xutil.c to this file. * libamu/xutil.c: explain purpose of file. Move mkdirs/rmdirs code from old util.c. @@ -555,22 +1129,13 @@ 2005-09-26 Erez Zadok <ezk@cs.sunysb.edu> - * libamu/xutil.c (unregister_amq): repair broken meaning of D_AMQ, - which was reversed in am-utils-6.2. Bug fix from Steve Plite - <splite-amutils at sigint.cs.purdue.edu>. - * hlfsd/stubs.c (nfsproc_getattr_2_svc, nfsproc_lookup_2_svc, nfsproc_readlink_2_svc, nfsproc_readdir_2_svc), hlfsd/hlfsd.c (hlfsd_init): use new clocktime() function. 2005-09-18 Erez Zadok <ezk@cs.sunysb.edu> - ******************************************************************* - *** Released am-utils-6.1.2.1 *** - ******************************************************************* - - * README.attrcache: Update status of freebsd and openbsd. - Document netbsd's patchs to support noac. + * README.attrcache: document netbsd's patchs to support noac. * libamu/mount_fs.c (compute_nfs_attrcache_flags): cleanup the code that sets the ac{reg,dir}{min,max} fields so it also sets the @@ -578,13 +1143,10 @@ * configure.in: check for nfs_args fields acregmax and acdirmax. - ******************************************************************* - *** Released am-utils-6.1.2 *** - ******************************************************************* - * README.attrcache: new file documenting in detail OS bugs relating to attribute caching, which can affect Amd's reliability under heavy load. + Update status of freebsd and openbsd. * doc/Makefile.am: install README.attrcache file as attrache.txt file on am-utils Web site. @@ -665,6 +1227,54 @@ amfs_bgmount, amfs_generic_mount_child), amd/amd.c (main): pass NULL to clocktime(). +2005-09-15 Erez Zadok <ezk@cs.sunysb.edu> + + * amd/nfs_subr.c (nfsproc_getattr_2_svc): initialize 'retry', just + in case. + (fh_to_mp3): increment amd_stats.d_stale counter only if we are + actually returning ESTALE. + + * amd/info_exec.c (exec_parse_qanswer), amd/info_file.c + (file_search_or_reload), amd/info_hesiod.c (hesiod_search), + amd/info_ldap.c (amu_ldap_search), amd/info_ndbm.c (search_ndbm), + amd/info_nis.c (nis_search), amd/info_nisplus.c (nisplus_search, + nisplus_search): don't dereference mnt_map->cfm which may be null + if we're not using an amd.conf file. + +2005-09-14 Erez Zadok <ezk@cs.sunysb.edu> + + * libamu/xutil.c (unregister_amq): if failed to de-register Amd + from portmapper, report it under dlog (debugging), not as an + annoying plog that always shows up. + +2005-09-13 Erez Zadok <ezk@cs.sunysb.edu> + + * mk-amd-map/mk-amd-map.c (read_file_file), libamu/util.c + (strsplit), libamu/hasmntopt.c (nextmntopt), amd/sun_map.c + (sun_entry2amd), amd/sun2amd.c (sun2amd_convert_convert), + amd/mapc.c (mapc_add_kv), amd/info_file.c (file_search_or_reload), + amd/info_exec.c (exec_parse_qanswer): cast isspace() and isascii() + arg to unsigned char, which is safer as per Christos. + + * amd/sun_map.c (sun_entry2amd): isspace takes an int, not a char. + + * configure.in: keep a dummy unused AM_PROG_LEX, because old + Automake (1.6.3) looks for it if you refer to LEX in your + Makefile.am files. Silly thing, fixed in newer versions of + Automake (1.9.2). + +2005-09-11 Erez Zadok <ezk@cs.sunysb.edu> + + * Makefile.am (EXTRA_DIST_M4): add prog_{lex,yacc}.m4 to distro. + + * configure.in: use new macros AMU_PROG_YACC and AMU_PROG_LEX. + + * m4/macros/prog_yacc.m4: new macro to check if bison/yacc/byacc + exist, but exist with an error if none found. + + * m4/macros/prog_lex.m4: new macro to check if f/lex exist, but + exit with an error if not found (am-utils requires f/lex). + 2005-09-07 Erez Zadok <ezk@cs.sunysb.edu> * minor new port: powerpc-apple-darwin8.2.0. @@ -683,8 +1293,48 @@ file system on this system (probably because /proc isn't mounted), then provide some default definition for this file to compile. +2005-08-28 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/sun_map.c (sun_entry2amd) : Wipe out any trailing white + spaces or '\n' before passing strings to the parser. + + * amd/amd.h : oops, reverting bad changes + + * amd/info_file.c : oops, reverting bad changes + + * amd/mapc.c : oops, reverting bad changes + + * amd/sun_map_parse.y : oops, reverting bad changes + 2005-08-27 Erez Zadok <ezk@cs.sunysb.edu> + * libamu/xutil.c: amd_program_number is a u_long now. + (unregister_amq): if pmap_unset failed to de-register Amd from the + portmapper, print an error message. + (get_amd_program_number): returns a u_long now. + (set_amd_program_number): take a u_long now. + + * include/am_utils.h: get_amd_program_number() returns u_long now. + set_amd_program_number() takes a u_long now. + + * amq/amq.c: amd_program_number is a u_long, not int. + + * amd/nfs_start.c (mount_automounter): when registering Amd with + the portmapper (for Amq service), don't fail if just one of TCP or + UDP failed to register, but only if both transports failed. That + way Amq can still contact Amd using the transport that did + register correctly. However, now, if both transports failed to + register, then set amd_state to "Done" so it would shutdown + cleanly. + + * conf/transp/transp_{sockets,tli{}.c (amu_svc_register): better + comments. + + * libamu/xutil.c: debug_flags should always be initialized to + D_CONTROL (amq,daemon,fork). + (debug_option): allow setting initial immutable debug_flags via + amd.conf; afterward, prevent it (i.e., one tries it via "amq -D"). + * Makefile.am (DISTCLEANFILES): also remove amu_autofs_prot.h, config.cache.old, and dbgcf.h on "make distclean". (CLEANFILES): remove dbgcf on plain "make clean". @@ -692,13 +1342,199 @@ * amd/Makefile.am (DISTCLEANFILES): also remove ops_autofs.c on "make distclean". +2005-08-27 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/info_file.c (file_search_or_reload): Changed the key + variable that is passed to the sun_entry2amd. + + * amd/mapc.c (mapc_add_kv): Add support for multiple entries + packed into one line. This is a workaround for handling Sun style + mounts that contains multiple entries on one line: "multi-mount + entries". The sun2amd conversion tools will convert such entries + into type:=auto and pack each of the auto-entries into one line + separated by '\n'. The mapc_add_kv function will now recognize + such entries and add then. + + * amd/sun2amd.c (sun2amd_convert): Add print statement that + includes the line number during a parser failure. + + * sun_map.c: Fix the AMD_MAP_PREF_KW definition. Fix spelling. + (sun_mountpts2amd): removed this function + (sun_hsfs2amd): added some more comments + (sun_nfs2amd): moved support for multi-mount entries from this + function to sun_multi2amd(). + (sun_multi2amd): New function to handle multi-mount entries. This + function will convert the Sun version to an Amd type:=auto. Each + extra auto entry will be appended to the same line separated by a + '\n'. + (sun_entry2amd): Fixed a bug if-statement. This function now + checks for multi-mount entries. + + * amd/sun_map.h: Add a fstype member to the sun_mountpt struct. + + * amd/sun_map_parse.y: Parser now supports fstype for multi-mount + entries. + + * amd/sun_map_tok.l: No longer print the line and column number + when parsing. This is because the parser is handed strings not + files. Therefore, the line number will always be 1. It is now up + to the higher level tools to echo line information during error + when they feed the parser. + 2005-08-24 Erez Zadok <ezk@cs.sunysb.edu> * configure.in: wrap all LDAP and HESIOD tests in test whether --with-OPT was used. +2005-08-20 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/sun_map.c: Clean up #defines + (sun_entry2amd) : Verify that the key and entry values passed are + not NULL. Free the Sun entry struct when done. + + * amd/sun_map.h: Add some "not implemented yet" comments to the + map and mmap structures. + + * amd/sun_map_parse.y: Removed automap include functionality, it + will need to be re-implemented using a pre-processor. + Clean up function calls. + +2005-08-16 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/sun2amd.c (sun2amd_convert): If the input line is too long + don't return an error just null terminate and continue. Fix bug: + the map key was not written. + + * amd/sun_map.c: Add a new Amd device, and cdfs constant. + (sun_locations2amd): Write the NFS type keyword + for each host of the mount entry. Add a space between each mount + location. + (sun_hsfs2amd): New function to support hsfs to + cdfs. + (sun_nfs2amd): Don't write the nfs type here. Fix + comments. + (sun_entry2amd): Change function parameter var for + clarity. Add support for hsfs. + + * amd/sun_map_parse.y (sun_map_parse_read): Reset the entry list + after it has been returned. + 2005-08-16 Erez Zadok <ezk@cs.sunysb.edu> + * amd/get_args.c (get_args): initialize debug_flags if they've + never been set. + +2005-08-15 Erez Zadok <ezk@cs.sunysb.edu> + + * amd/mapc.c (mapc_create): mapc find takes 4th arg, the mount + point. Now, we can properly initialize the cfm structure of the + mnt_map, so we can have access to the flags (e.g., SUN_MAP_SYNTAX) + inside each info_*.c file. + (mapc_find): takes 4rd arg (mntpt) and passes it to calls to + mapc_create. + (root_newmap): pass NULL as 4th arg (mntpt). + + * amd/amfs_generic.c (amfs_parse_defaults): remove ugly hack to + initialize map cfm structure (including flags) so late. + + * amd/autil.c (amfs_mkcacheref): pass mount point as 4th arg to + mapc_find. + + * amd/amfs_root.c (amfs_root_mount): pass NULL as new 4th arg to + mapc_find. + + * amd/amd.h: extern for mapc_find changed to pass the mount point. + + * libamu/xutil.c: debug_flags should be unsigned. + (dbg_opt): fix comments for updated debug_flags, add "defaults". + (debug_option): Don't allow "immutable" (D_IMMUTABLE) flags to be + changed, because they could mess Amd's state and only make sense + to be set once when Amd starts. Currently these immutable flags + are "daemon," "fork," "amq," and "mtab." If Amd not compiled with + debugging, set default "control" options (daemon,fork,amq). + (unregister_amq): proper use of D_AMQ. + + * include/am_utils.h: define new D_BASIC, D_CONTROL, D_DEFAULT, + D_IMMUTABLE, and D_MASK. Updated definitions of D_ALL and D_TEST. + debug_flags extern should be unsigned. + + * scripts/amd.conf-sample (debug_options): mention new "defaults" + option. + + * scripts/amd.conf.5,doc/am-utils.texi: document new debug option + "defaults" and properly explain the corrected meanings of + "daemon," "fork," "amq," and their "no" inversions. Better + explain "all" and "test" too. Correct default debug_mtab file to + /tmp/mtab. + + * hlfsd/stubs.c (nfsproc_readlink_2_svc): proper use of D_FORK. + + * hlfsd/homedir.c (homedir): proper use of D_FORK. + + * hlfsd/hlfsd.c (main, hlfsd_init, cleanup): proper use of + D_DAEMON. + + * amd/nfs_start.c (mount_automounter): proper use of D_AMQ. + + * amd/amd.h (DEBUG_MNTTAB_FILE): change default to /tmp/mtab, + which is more common on more systems (instead of /tmp/mnttab). + + * amd/amd.c (main): properly use D_DAEMON. + +2005-08-14 Erez Zadok <ezk@cs.sunysb.edu> + + * libamu/xutil.c (xlog_opt): define new pseudo log_option named + "defaults", which is an alias for "fatal,error,user,warning,info". + + * scripts/amd.conf.5, doc/am-utils.texi, amd/amd.8: document new + log_option "defaults". + + * amq/amq.8: typo. Arg name for -D is "debug_options" not + "log_options". + + * libamu/xutil.c: make xlog_level static, and initialize it to + XLOG_DEFAULT. + (cmdoption): take unsigned int as flags argument, and print any + errors as XLOG_ERROR (which cannot be turned off). + (switch_option): remove old code which disallowed you from turning + off flags that were set when Amd started up. Now, you can turn + any log option on/off at start time or later on (via amq -x), + other than the two mandatory options (fatal+error). + + * include/am_utils.h (XLOG_*): remove unused XLOG_DEFSTR. Define + XLOG_MASK to mask out unused flag bits. Define XLOG_DEFAULT to be + what it was before (fatal + error + user + warning + info), but + use the new mask. Remove extern's for xlog_level and + xlog_level_init. Change extern for cmdoption() to take an + unsigned flags argument. + + * doc/am-utils.texi, amq/amq.8, scripts/amd.conf.5, amd/amd.8: + document mandatory log options + (fatal+error) which cannot be turned off and are always on by + default. They must be on so that Amd can report crucial errors, + including those related to setting flags on/off. + + * hlfsd/hlfsd.c (main): remove hack to initialize log_options. + + * wire-test/wire-test.c: remove reference to unused xlog_level. + + * amd/mapc.c (mapc_create): 'alloc' flags variable should be + unsigned. + + * amd/get_args.c (get_args): no need for hack using + xlog_level_init. + + * amd/Makefile.am: separate sources common to amd and sun2amd into + a special, locally-built only, static library called libaminfo.a. + + * libamu/xutil.c (real_plog, xsnprintf): call new xvsnprintf() + wrapper function. + (xvsnprintf): new function which already gets a va_list. This was + needed to avoid nesting va_list's (which apparently isn't + allowed). + + * include/am_utils.h: prototype for new xvsnprintf wrapper. + * configure.in: sinclude([vers.m4]) version number file. * vers.m4: new file to separate version number out of @@ -707,24 +1543,248 @@ * Makefile.am (EXTRA_DIST_M4): include vers.m4 in distro. +2005-08-14 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/info_file.c: add key to sun_entry2amd()'s parameters + + * amd/info_hesiod.c: add key to sun_entry2amd()'s parameters + + * amd/info_ldap.c: add key to sun_entry2amd()'s parameters + + * amd/info_ndbm.c: add key to sun_entry2amd()'s parameters + + * amd/info_nis.c: add key to sun_entry2amd()'s parameters + + * amd/info_nisplus.c: add key to sun_entry2amd()'s parameters + + * amd/sun_map.c: New functions to support sun translation to amd. + The function sun_entry2amd's parameters were expanded to include + the key value. sun_entry_2amd() now returns NULL on error. + + * amd/sun_map.h: Expand sun_entry2amd's parameters to include key + value. + + * amd/sun_map_parse.y: Change how file system types are parsed out + of the list of options for each entry. The previous way was not + working properly so now we just look for the 'fstype' keyword in + each option. If it exists we copy its key value. + Added some comments to sun_map_parse_read(). + + * amd/amd.h: Added an extern from info_file.c called + file_read_line(). Previously this function was static but is now + used in sun2amd.c. + + * amd/info_file.c: Changed the static function read_line() to a + public function called file_read_line(). This function is now + used in sun2amd.c + + * amd/sun2amd.8: Filled in the contents of this man page. This is + a conversion utility that converts Sun maps to Amd maps. + + * amd/sun2amd.c: Implemented a working version of this utility. + 2005-08-11 Erez Zadok <ezk@cs.sunysb.edu> + * amd/sun_map_parse.y: instead of using a custom FOO_alloc() + function, use Amd's CALLOC(struct FOO) macro. + + * amd/sun_map.h: remove extern definitions to functions no longer + needed. + + * amd/sun_map.c: removed superfluous *_alloc() functions, some of + which caused conflicts with same-named symbols in the parser + (since we rename 'yyalloc' to 'sun_map_alloc'). + * configure.in: keep AC_INIT on same line, for nightly snapshot build script. + * amd/Makefile.am: f/lex libraries are incompatible with multiple + scanners, so don't link Amd with libl.a/libfl.a. + +2005-08-10 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/sun_map_tok.l: Clean up regular expression so HPUX stops + whining. + +2005-08-09 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/sun_map_parse.y: Moved license below special yacc bracket to + be portable with HPUX's yacc. Clean up externs. + + * amd/sun_map_tok.l: Moved license below special lex bracket to be + portable with HPUX's lex. Fixed definitions to allow for a + non-flex lex to parse strings instead of files. Added some casts + to strlcpy usage to silence warnings. Moved the % options because + HPUX was complaining, + +2005-08-08 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/sun_map_tok.l: Applied some definition goop to handle the + ECHO symbol that lex defines. This symbol causes problems on + RedHat-EL-powerPC platforms. Replaced strncpy with strlcpy. + Renamed the function sun_map_yyinput to sun_map_input. This + function is a utility function that is not generated by lex. + Therefore, the 'yy' may cause confusion. + +2005-08-08 Erez Zadok <ezk@cs.sunysb.edu> + + * amd/sun_map_tok.l: allocate more output slots so lex scanners + don't run out of memory. + + * amd/sun_map_parse.y: declare extern for sun_map_parse(). + + * libamu/xutil.c (xsnprintf): ensure proper comparisons between + signed and unsigned quantities. + +2005-08-06 Erez Zadok <ezk@cs.sunysb.edu> + + * libamu/util.c, amd/autil.c (strsplit): moved strsplit() function + from Amd code to to libamu, so sun2amd could use it. + + * amd/Makefile.am (sbin_PROGRAMS): build sun2amd binary and man + page. + (EXTRA_sun2amd_OBJECTS): sun2amd needs to link with all info_*.c files + + * amd/sun2amd.8: placeholder for man page. + + * amd/sun2amd.c: placeholder for C version of translator from Sun + syntax maps to Amd maps. + + * */*.[0-9]: include file name in nroff source comment. + + * ALL: use '0' properly when assigning or passing it to functions. + If in the context pointer, use NULL. If in the context of a + single char (say within a string), the use '\0'. This is just to + make the code clearer. + + * m4/macros/header_templates.m4: template for HAVE_MAP_SUN. + + * amd/mapc.c (maptypes): add placeholder for Sun-syntax map + methods. + + * amd/Makefile.am (EXTRA_amd_SOURCES): compile info_sun.c + + * amd/info_sun.c: placeholder for meta info parser to get info + from Sun automount-style /etc/auto_master, possibly following + into multiple info services (via /etc/nsswitch.conf). + + * libamu/xutil.c (xsnprintf): if vsnprintf truncated the output + string to avoid an overflow, print an error. Include some code to + break out any possible infinite loop between plog() and + xsnprintf(). + (real_plog): now we can use (carefully) xsnprintf() directly. + + * amd/sun_map.[hc]: cleanup and formatting. + + * amd/sun_map_parse.y: to match the literal string "fstype=" use + double quotes, not single quotes. + + * amd/get_args.c (get_args): call conf_parse() using conf_in. + + * amd/amd.h: use renamed parser symbols (yyparse -> conf_parse; + yyin -> conf_in). + + * ylwrap: Enhance wrapper to avoid the #define hacks done by Gdb, + and suggested by the Automake manual. This enhanced parser will + prefix each conflicting yacc/lex symbol with a unique name + produced from INPUT. For example, if INPUT is foo_parse.y, then + yyparse() becomes foo_parse(). This requires that both your .l + and corresponding .y files have the same prefix (e.g., foo_parse.y + and foo_tok.l). + + * amd/Makefile.am (BUILT_SOURCES): define which yacc/lex .h/.c + source files are built automatically, and therefore should not be + in the distro (IMHO), and should be cleaned upon "make clean". + (amd_SOURCES): include new sun_map_parse.y and sun_map_tok.l files. + + * ylwrap: new file needed because we have two parsers now. + + * amd/sun_map_{parse.y,tok.l}: cleanup and formatting. + +2005-08-06 Daniel P. Ottavio <dottavio@ic.sunysb.edu> + + * amd/Makefile.am: Added new files sun_map.h, sun_map.c, + sun_map_parse.y, and sun_map_tok.l. These files add support for + parsing Sun style maps. The yacc/lex files needed custom rules so + they don't cause symbol conflicts with each other. + + * amd/amd.h: Added "#define INFO_MAX_LINE_LEN" that specifies the + line limit for info services. + + * amd/info_exec.c: Replaced MAX_LINE_LEN with INFO_MAX_LINE_LEN + + * amd/info_file.c: Replaced MAX_LINE_LEN with INFO_MAX_LINE_LEN, + added support for parsing Sun style maps. The support for Sun + maps is done using the sun_entry2amd() function. This function is + called for each entry line that is read only when Sun parsing is + enabled. For each entry line that is passed to the function a + copy of the Amd equivalent is returned. + + * amd/sun_map.h: new header file for Sun style map support + + * amd/sun_map.c: new source file for Sun style map support, clean + up comments, additional sun2amd conversion routines + + * amd/sun_map_parse.y: new yacc file for parsing Sun style maps, + fix copyright notice, put filename below license. + + * amd/sun_map_tok.l: new lex file for parsing Sun style maps, fix + copyright notice, put filename below license. + + * amd/get_args.c: replace usage of strlcat with xstrlcat + + * include/am_utils.h: added 'extern void xstrlcat()' + + * libamu/util.c: Added a xstrlcat() function that wraps the + syscall strlcat() so that string truncations can be logged. + 2005-08-06 Erez Zadok <ezk@cs.sunysb.edu> + * amd/info_file.c (file_search, file_reload): pass mnt_map to + file_search_or_reload. + (file_search_or_reload): ensure that mnt_map is never NULL. + + * amd/info_nis.c (nis_search): call sun_entry2amd() if + sun_map_syntax=yes. + + * amd/info_ldap.c (amu_ldap_search): call sun_entry2amd() if + sun_map_syntax=yes. + + * amd/info_hesiod.c (hesiod_search): call sun_entry2amd() if + sun_map_syntax=yes. + + * amd/info_ndbm.c (ndbm_search): pass mnt_map to search_ndbm. + (search_ndbm): call sun_entry2amd() if sun_map_syntax=yes. + + * amd/info_nisplus.c (nisplus_search): call sun_entry2amd() if + sun_map_syntax=yes. + + * amd/info_exec.c (exec_search): pass mnt_map to + exec_parse_qanswer. + (exec_parse_qanswer): call sun_entry2amd() if sun_map_syntax=yes. + + * doc/am-utils.texi (sun_map_syntax Parameter): document new + common parameter. + * scripts/amd.conf.5: selectors_in_defaults is a common parameter, not just [global]. Use consistent capitalization of Amd/Amq. + Document new sun_map_syntax parameter. * scripts/amd.conf-sample: properly list all of the parameters which are common to both the [global] and the per-map sections. + (sun_map_syntax): example of new parameter. * doc/am-utils.texi (Common Parameters): selectors_in_defaults is a common parameter, not just [global]. * scripts/amd.conf-sample (sun_map_syntax): example of new flag. + * amd/conf.c ({ropt,gopt}_sun_map_syntax): new function to parse + sun_map_syntax flag (global or per map). + + * amd/amd.h (CFM_SUN_MAP_SYNTAX): new flag for users to say if the + map uses Sun automounter syntax. + * cvs-server.txt: update instructions after branching 6.1-stable. 2005-08-02 Erez Zadok <ezk@cs.sunysb.edu> |